Large Language Models in Ophthalmology: Potential and Pitfalls.

Large language models (LLMs) show great promise in assisting clinicians in general, and ophthalmology in particular, through knowledge synthesis, decision support, accelerating research, enhancing education, and improving patient interactions. Specifically, LLMs can rapidly summarize the latest literature to keep clinicians up-to-date. They can also analyze patient data to highlight crucial insights and recommend appropriate tests or referrals. LLMs can automate tedious research tasks like data cleaning and literature reviews. As AI tutors, LLMs can fill knowledge gaps and assess competency in trainees. As chatbots, they can provide empathetic, personalized responses to patient inquiries and improve satisfaction. The visual capabilities of LLMs like GPT-4 allow assisting the visually impaired by describing environments. However, there are significant ethical, technical, and legal challenges around the use of LLMs that should be addressed regarding privacy, fairness, robustness, attribution, and regulation. Ongoing oversight and refinement of models is critical to realize benefits while minimizing risks and upholding responsible AI principles. If carefully implemented, LLMs hold immense potential to push the boundaries of care, discovery, and quality of life for ophthalmology patients.

Understanding choroidal nevus risk factors for transformation into melanoma.

A choroidal nevus is a common intraocular tumor in the United States, found in approximately 5% of Caucasian adults. The three main risks of melanocytic choroidal nevus include vision loss from a subfoveal nevus, development of subretinal fluid, and transformation of nevus into melanoma, a malignant counterpart. We explore clinical risk factors that predict benign melanocytic choroidal nevus transformation into a malignant choroidal melanoma. Based on a large analysis of 2,355 cases that were monitored longitudinally using multimodal imaging, the most recent list of clinical features includes tumor Thickness greater than 2 mm on ultrasonography, subretinal Fluid on optical coherence tomography, Symptomatic vision loss 20/50 or worse, Orange pigment on fundus autofluorescence, Melanoma hollow on ultrasonography, and DIaMeter greater than 5 mm on fundus photography. These factors are remembered with a mnemonic of the capital letters TFSOM-DIM for "To Find Small Ocular Melanoma Doing Imaging." Analysis of these factors demonstrated a Kaplan-Meier mean five-year risk of 1% with no risk factors, 11% with any one factor, 22% with any two factors, 34% with any three factors, 51% with any four factors, and 55% with any five factors. There was no patient with six risk factors. Of those with combinations of four risk factors, six of 15 combinations yielded a 70%-100% rate of transformation; of those with combinations of five risk factors, two of five combinations yielded a 70%-100% rate of transformation. Choroidal nevus carries a risk for evolving into melanoma, and understanding of clinical and imaging features predictive of this outcome is highly important.

Conjunctival melanoma: Insights into classification, outcomes, and biomarkers.

Conjunctival melanoma is quite rare, estimated at approximately 0.5 incidence per 1 million persons per year. This malignancy arises from a pre-existing nevus (7%), primary acquired melanosis (74%), or de novo without pre-existing condition (19%) and develops most often in patients with Fitzpatrick skin types I (23%) and II (62%). At initial presentation, the tumor size is approximately 13 mm in cross-sectional diameter and has 3-mm thickness, involving the bulbar (97%), forniceal (30%), tarsal (28%), or caruncular (11%) regions, often with corneal (54%) and rarely with orbital (4%) involvement. According to the eighth edition of the American Joint Committee on Cancer (AJCC), the tumor is classified as T1 (63%), T2 (18%), T3 (20%), and T4 (0%). Outcomes depend on several factors including patient age, AJCC classification, orbital invasion, and type of initial surgery, whereas tumor origin and Fitzpatrick skin type do not appear to impact outcomes. Older patients (≥70 years of age) demonstrate larger tumors, greater recurrence, and greater vision loss. Analysis of 425 patients by AJCC classification (T1 versus T2 versus T3) revealed increasing T category with greater lymph node metastasis (3% versus 13% versus 25%; P < .001), tumor-related systemic metastasis (13% versus 45% versus 40%; P < .001), and tumor-related death (8% versus 22% versus 37%; P < .001). Data of patients with orbital invasion revealed significantly greater 10-year rates of exenteration (P < .001), distant metastasis (P = .0005), and death (P = .001). Studies have demonstrated biomarkers related to conjunctival melanoma include mutations in BRAF, NRAS, ATRX, and NF1. Future therapies might be directed against these mutations or with small-molecule inhibitors and/or immunotherapy. In summary, conjunctival melanoma is a rare but ominous malignancy, imparting moderate risk for lymph node and systemic metastasis as well as death, depending on tumor features and classification. The first surgery is highly important in prevention of tumor seeding, recurrence, and metastasis.

Unilateral acute iris transillumination syndrome with glaucoma and iris pigment epithelium dispersion simulating iris melanoma.

Purpose: To report a patient with a unilateral presentation of glaucoma, pain, and acute iris transillumination syndrome simulating iris melanoma. Observations: A 53-year-old male presented with blurred vision and pain in his right eye several weeks following a respiratory sinus infection managed by oral azithromycin. Examination of the right eye was notable for elevated intraocular pressure of 46 mm Hg, an irregular mid-dilated pupil, and diffuse iris transillumination with pigmentary seeding on the iris surface, in the anterior chamber angle, and on the sclera, suspicious for diffuse iris melanoma with glaucoma and extrascleral extension. Ultrasound biomicroscopy (UBM) of the right eye revealed circumferential anterior chamber angle and trabecular meshwork involvement by an infiltrative process corresponding to the pigmented cells noted clinically, while the ciliary body was unremarkable. Following enucleation, histopathology showed extensive necrosis of the iris pigment epithelium, sphincter, and dilator muscles with melanophagic infiltration in the anterior chamber angle and episclera, mild chronic non-granulomatous iridocyclitis, and no evidence of a melanocytic neoplasm. Although immunohistochemical studies for herpes simplex virus (HSV) types 1 and 2, varicella-zoster virus, and cytomegalovirus were negative, qualitative real-time polymerase chain reaction on paraffin-embedded tissue detected HSV-1 DNA. The combined clinical, pathologic, and molecular findings were compatible with unilateral acute iris transillumination syndrome, likely HSV-1 associated. Conclusion and Importance: Unilateral acute iris transillumination syndrome with diffuse iris pigment epithelial loss can simulate iris melanoma. Prompt herpes viral studies may be informative.

Conditional Metastasis of Uveal Melanoma in 8091 Patients over Half-Century (51 Years) by Age Group: Assessing the Entire Population and the Extremes of Age.

Purpose: To evaluate cumulative incidence of metastasis at specific timepoints after treatment of uveal melanoma in a large cohort of patients and to provide comparison of conditional outcomes in the youngest and oldest cohorts (extremes of age). Methods: Retrospective analysis of 8091 consecutive patients with uveal melanoma at a single center over a 51-year period. The patients were categorized by age at presentation (0-29 years [n = 348, 4%], 30-59 years [n = 3859, 48%], 60-79 years [n = 3425, 42%], 80 to 99 years [n = 459, 6%]) and evaluated for nonconditional (from presentation date) and conditional (from specific timepoints after presentation) cumulative incidence of metastasis at five, 10, 20, and 30 years. Results: For the entire population of 8091 patients, five-year/10-year/20-year/30-year nonconditional cumulative incidence of metastasis was 15%/23%/32%/36%, and the conditional incidence improved to 6%/15%/25%/30% for patients who did not develop metastasis in the first three years. For the extremes of age (0-29 years and 80-99 years), the nonconditional cumulative incidence of metastasis revealed the younger cohort with superior outcomes at 8%/15%/19%/27% and 21%/29%/29%/29%, respectively (P < 0.001). The conditional incidence (at one-year and two-year timepoints with metastasis-free survival) showed persistent superior younger cohort survival (P < 0.001, P = 0.001), but no further benefit for patients with three-year metastasis-free survival at 4%/12%/16%/24% and 7%/18%/18%/18%, respectively (P = 0.09). Conclusions: Non-conditional metastasis-free survival analysis for patients with uveal melanoma revealed the youngest cohort to have significantly better survival than the oldest cohort, and this persisted into one-year and two-year conditional metastasis-free survival but diminished at the three-year conditional timepoint.

Conjunctival Melanoma in 430 Cases: Comparative Analysis of the Impact of Orbital Invasion on Tumor Recurrence, Metastasis, and Death.

PURPOSE: To compare the clinical features at presentation and treatment outcomes of conjunctival melanoma by absence/presence of orbital invasion. METHODS: A retrospective review of patients with conjunctival melanoma managed at a single tertiary referral center from April 18, 1974, to September 9, 2019. RESULTS: Of 430 patients with conjunctival melanoma, 21 (5%) had orbital invasion at presentation. A comparison between the 2 groups (orbital invasion absent vs. present) revealed that the orbital invasion group had a higher frequency of prior eyelid incisional biopsy (5% vs. 24%, P = 0.006), greater tumor basal diameter (12.2 vs. 17.3, P = 0.009), greater tumor thickness (2.4 vs. 7.0, P < 0.001), more quadrants involved (1.8 vs. 2.5, P = 0.002), and more clock hours involved (4.4 vs. 5.8, P = 0.037). In addition, those with orbital invasion were more likely to undergo exenteration as primary treatment (1% vs. 24%, P < 0.001). Multivariate relative risk regression analysis revealed that variables predictive of orbital invasion included greater tumor thickness (P < 0.001) and greater involvement of the fornix (P = 0.031) and tarsus (P = 0.033). Outcomes revealed orbital invasion group with greater 5-year/10-year distant metastatic rate (16%/21% vs. 63%/63%, P = 0.005), and greater melanoma-related death rate (7%/13% vs. 38%/53%, P = 0.001). CONCLUSIONS: Conjunctival melanoma with orbital invasion at presentation demonstrate larger, more extensive tumors involving the fornix or tarsus, and with greater rate of melanoma-related metastasis and death.

Retinoblastoma: emerging concepts in genetics, global disease burden, chemotherapy outcomes, and psychological impact.

In this review we discuss several recent concepts regarding retinoblastoma control and its impact. In a cohort of 482 patients with solitary unilateral retinoblastoma revealed germline mutation in 16% and the likelihood of germline retinoblastoma was greater for younger children (≤1 year versus (vs.) >1 year at presentation) with odds ratio (OR) 2.96 (p = 0.001), and greatest for the youngest infants (≤3 months vs. >3-12 months) (OR 5.52) (p = 0.002). Retinocytoma/retinoma, a benign variant of retinoblastoma, was studied in 78 tumours and demonstrated transformation into retinoblastoma in 9.2% by 5 years and 15.3% by 10 years and 20 years. An international global study on retinoblastoma over 1.5 years revealed 4351 new patients and 85% from low- and middle-income countries, notably with older age at detection and greater risk for metastasis. Management of retinoblastoma in 964 eyes using intravenous chemotherapy showed 20-year globe salvage at 96% in group A, 90% in group B, 90% in group C, 68% in group D, and 32% in group E eyes. The 5-year globe salvage with intra-arterial chemotherapy for 160 eyes (655 infusions) with retinoblastoma showed success in 100% for group B, 80% for group C, 78% for group D, and 55% for group E. The psychological impact of retinoblastoma on the parents revealed depression (73%), anxiety (64%), and/or stress (100%), and on the patient revealed deficits in quality of life issues. Retinoblastoma is a challenging disease and chemotherapy provides reliable tumour control and globe salvage. Continuing efforts to improve quality of life issues is important.

Metastatic tumours to the eye. Review of metastasis to the iris, ciliary body, choroid, retina, optic disc, vitreous, and/or lens capsule.

Metastasis to the eye can involve the choroid (90%), ciliary body (2%), iris (8%), and retina, optic disc, vitreous, and/or lens capsule (<1-4%). The mean number of uveal metastasis per eye (1.7), mean tumour base (11.6 mm) and thickness (3.2 mm), tumour colour (86% yellow), and presence of subretinal fluid (72%), are all clinical features suggestive of the diagnosis. Imaging with ultrasonography demonstrates an echodense mass (80%) and optical coherence tomography shows a "lumpy bumpy" choroidal surface (64%), both important diagnostic features. Uveal metastases typically emanate from primary cancer of the breast (37%), lung (27%), kidney (4%), gastrointestinal tract (4%), cutaneous melanoma (2%), lung carcinoid (2%), prostate (2%), thyroid (1%), pancreas (1%), and other sites (3%). Occasionally, fine needle aspiration biopsy is employed if the primary site is not known. In 16% of cases, the primary site remains unknown. Rarely, metastases affect the retina, vitreous, and lens capsule, most often originating from cutaneous melanoma and in patients previously treated with checkpoint inhibitor therapy. Kaplan-Meier analysis in a series of 1111 patients with uveal metastasis revealed 32% survival at 3 years and 24% at 5 years. Patients with uveal metastasis from carcinoid tumour showed most favourable survival at 5-years (92%), whereas pancreatic and kidney cancer demonstrated least favourable survival (0%). The 5-year survival was better for females (versus (vs.) males) (31% vs. 21%) and older adults (vs. children) (40% vs. 0%). In this review, we examine several large-cohort publications on the topic of ocular metastasis.

Iris melanoma: factors predictive of post-management secondary glaucoma in 271 cases at a Single Ocular Oncology Centre.

BACKGROUND/AIMS: To identify factors predictive of post-management secondary glaucoma in eyes with iris melanoma. METHODS: Patients with iris melanoma who were conservatively managed on the Ocular Oncology Service, Wills Eye Hospital from 1970 to 2016 were included. Charts were retrospectively reviewed, and binary logistic regression analysis was performed. Main outcome measures were factors predictive of post-management glaucoma, defined as intraocular pressure (IOP) > 22 mmHg following melanoma treatment. RESULTS: Of 271 patients with iris melanoma, melanoma-related glaucoma was identified in 40 (15%) at presentation and post-management glaucoma developed in 75 (28%) at a mean of 103.7 months (range:1.0-120.0). Comparison (post-management glaucoma vs. no glaucoma) revealed patients with post-management glaucoma presented with worse visual acuity (20/50-20/150) (17% vs. 5%, p = 0.001), increased mean tumour basal diameter (5.1 mm vs. 4.3 mm, p = 0.004), greater melanoma-related increased IOP on presentation (24.1 mmHg vs. 16.2 mmHg, p < 0.001), diffuse tumour shape (9% vs. 4%, p = 0.01), American Joint Committee on Cancer (AJCC) T4 category (7% vs. 2%, p = 0.03), and extraocular tumour extension (7% vs. 2%, p = 0.03). Risk factors for post-management glaucoma identified by multivariate analysis included melanoma-related increased IOP at presentation (OR:1.1, [1.08-1.22] per 1-mmHg increase, p < 0.001), increased mean tumour basal diameter (OR:1.17, [1.02-1.33] per 1-millimetre increments, p = 0.03), advanced AJCC clinical T subcategory (OR:1.23, [1.04-1.46] per 1-subcategory increments, p = 0.02) and plaque radiotherapy treatment (OR:2.32, [1.13-4.75], p = 0.02). CONCLUSION: Features of iris melanoma that predicted post-management glaucoma included melanoma-related increased IOP on presentation, advanced AJCC clinical T subcategory, increased mean tumour basal diameter, and plaque radiotherapy treatment.

Choroidal Melanoma Masquerading as Central Serous Chorioretinopathy.

PURPOSE: To determine the clinical features and outcomes of choroidal melanoma initially masquerading as central serous chorioretinopathy (CSCR). DESIGN: Retrospective case series. SUBJECTS: All patients with choroidal melanoma, initially misdiagnosed as CSCR elsewhere and evaluated by the Ocular Oncology Service at Wills Eye Hospital from 2004 to 2022, were included. METHODS: A retrospective detailed review of patient charts and imaging was performed for all patients included in the study. Paired t tests and chi-squared tests were performed for data analysis. MAIN OUTCOME MEASURES: The primary outcome measures included clinical characteristics, ultrasonography, OCT, fundus autofluorescence, fluorescein angiography, and indocyanine green angiography. The secondary outcome measures included treatment results, such as the final visual acuity, tumor control, radiation-related complications, and melanoma-related metastlasis and death. RESULTS: There were 22 patients (mean age, 48 years; 16 men) in this cohort. The mean interval between initial CSCR diagnosis and suspicion of choroidal melanoma was 50 months (median, 50 months; range, 0-242 months). At tumor diagnosis, the melanoma was submacular in 16 (73%) patients and extramacular in 6 (27%) patients. The mean tumor thickness was 3.4 mm (median, 2.5 mm; range, 1.4-10.7 mm), and the mean basal diameter was 9.2 mm (median, 8.0 mm, range, 4.5-22.0 mm). Features enabling differentiation of choroidal melanoma from CSCR (affected versus unaffected eye) included choroidal thickness asymmetry (100% > 300 µm versus 21% > 300 µm; P = 0.005), ipsilateral choroidal surface irregularity (100% versus 0%; P < 0.001), loss of choroidal vascular detail on OCT (100% versus 0%; P < 0.001), presence of multiple pinpoint leaks on angiography (100% versus 0%; P < 0.001), and contralateral lack of autofluoresence abnormalities (75% versus 6%; P = 0.001). Management of the choroidal melanoma included plaque radiotherapy (19, 86%), enucleation (2, 9%), or treatment elsewhere (1, 5%). On follow-up (mean, 6 years), vision loss of ≥ 3 Snellen lines (9 patients, 47%), metastasis (3 patients, 14%), and death (1 patient, 5%) were noted. CONCLUSIONS: Patients with presumed CSCR, especially if chronic, should be evaluated for a possible thin underlying choroidal melanoma with a dilated fundus examination and multimodal imaging.

Likelihood of germline mutation with solitary retinoblastoma based on tumour location at presentation.

BACKGROUND/AIMS: To evaluate the likelihood of germline mutation in patients presenting with solitary retinoblastoma based on tumour location at first examination. METHODS: Retrospective analysis of solitary unilateral retinoblastoma for likelihood of germline mutation (family history of retinoblastoma and/or genetic testing indicating germline RB1 mutation and/or development of additional new or bilateral tumours) based on tumur location at presentation (macular vs extramacular). RESULTS: Of 480 consecutive patients with solitary retinoblastoma, 85 were in the macula (18%) and 395 were extramacular (82%). By comparison (macular vs extramacular tumours), macular tumours had smaller basal diameter (12.7 mm vs 18.9 mm, p<0.001) and smaller tumour thickness (6.1 mm vs 10.7 mm, p<0.001). Patients with macular tumours demonstrated greater likelihood for germline mutation (23% vs 12%, OR=2.18, p=0.011), specifically based on family history of retinoblastoma (13% vs 2%, OR=4.64, p=0.004), genetic testing showing germline RB1 mutation (27% vs 15%, OR=2.04 (95% CI 1.04 to 4.01), p=0.039), development of new tumours (13% vs 3%, OR=5.16 (95% CI 2.06 to 12.87), p=0.001) and/or development of bilateral disease (9% vs 2%, OR=4.98 (95% CI 1.70 to 14.65), p=0.004). CONCLUSIONS: Among patients with solitary unilateral retinoblastoma, those presenting with macular tumour (compared with extramacular tumour) show 2.18 times greater likelihood for germline mutation and an even higher likelihood of development of subsequent tumours. Solitary macular retinoblastoma should raise an index of suspicion for likely germline mutation and multifocal disease.

Conditional survival of uveal melanoma using The Cancer Genome Atlas (TCGA) classification (Simplified Version) in 1001 cases.

PURPOSE: To understand conditional prognostic value of the Cancer Genome Atlas (TCGA) for uveal melanoma metastasis based on event-free survival at 1, 2, 3, 4, and 5 years. METHODS: A retrospective study of eyes with uveal melanoma categorized according to TCGA and studied for nonconditional and conditional risks for metastasis at 5 and 10 years. RESULTS: Of 1001 eyes with uveal melanoma, the nonconditional (standard, at presentation) 5-year/10-year metastatic rate was 18%/25%. The conditional 5-year/10-year metastatic rate (for those without metastasis at 2 years) revealed 10%/18% and the conditional 10-year metastatic rate (for those without metastasis at 5 years) revealed 9%. The TCGA categories included Group A (n = 486, 49%), B (n = 141, 14%), C (n = 260, 26%), and D (n = 114, 11%). The non-conditional 5-year/10-year metastatic rate revealed Group A (4%/6%), Group B (12%/20%), Group C (23%/49%), and Group D (60%/68%). The conditional 5-year/10-year metastatic rate (for those without metastasis at 2 years) revealed Group A (2%/5%), Group B (8%/18%), Group C (21%/40%), and Group D (38%/50%). The conditional 10-year metastatic rate (for those without metastasis at 5 years) revealed Group A (2%), Group B (10%), Group C (33%), and Group D (20%). The peak incidence of metastasis for Groups A and B occurred during years 5-6, C during years 4-6, and D during years 1-2. CONCLUSION: Survival outcomes for uveal melanoma as non-conditional (at presentation) and conditional (event-free survival during follow-up) reveal reduction in metastatic rate over time. For those with 5-year metastasis-free survival, the 10-year conditional risk for metastasis was 9%.

Mutational Landscape and Outcomes of Conjunctival Melanoma in 101 Patients.

PURPOSE: To evaluate targetable mutations and molecular genetic pathways in conjunctival melanoma with clinical correlation. DESIGN: Observational case series. PARTICIPANTS: Patients with conjunctival melanoma. MAIN OUTCOME MEASURES: Mutational profile of the tumor by next-generation sequencing (NGS), alternative lengthening of telomeres (ALT) by fluorescence in situ hybridization (FISH), and ATRX immunohistochemistry. Outcomes at 2 years and 5 years of tumor-related metastasis and death were recorded. RESULTS: Of the 101 patients, mean age at presentation was 60 years, 52% were male, and 88% were White. The NGS panels initially targeted BRAF only (n = 6, 6%), BRAF/NRAS (n = 17, 17%), and BRAF/NRAS/NF1 (n = 10, 10%). Sixty-eight tumors were tested with the expanded 592-gene panel. Next-generation sequencing identified high-frequency mutations in NF1 (29/74, 39%), BRAF (31/101, 31%), NRAS (25/95, 26%), and ATRX (17/68, 25%). Of those with an ATRX mutation, 12 (71%) had an additional NF1 mutation. A subset analysis of 21 melanomas showed that the ATRX mutation was associated with loss of ATRX protein expression and ALT. Loss of ATRX expression and ALT were present in both intraepithelial and invasive tumors, suggesting that an ATRX mutation is an early event in conjunctival melanoma progression. The NF1 and ATRX mutations were associated with tarsal (vs. nontarsal) tumors (NF1: 28% vs. 9%, P = 0.035, ATRX: 41% vs. 14%, P = 0.021) and orbital (vs. nonorbital) tumors (ATRX: 24% vs. 2%, P = 0.007). ATRXMUT (vs. ATRXWT) tumors were associated with a lower 2-year rate of metastasis (0% vs. 24%, P = 0.005). NRASMUT (vs. NRASWT) tumors were associated with a greater 2-year rate of metastasis (28% vs. 14%, P = 0.07) and death (16% vs. 4%, P = 0.04), with a 5-fold increased risk of death (relative risk, 5.45 [95% confidence interval, 1.11-26.71], P = 0.039). CONCLUSIONS: This study confirms the high frequency of previously documented BRAF and NRAS mutations and recently reported ATRX and NF1 mutations in conjunctival melanoma. An NRAS mutation implied increased risk for metastasis and death. Loss of ATRX and ALT may be early events in conjunctival melanoma development.

Multimodal imaging of chorioretinal folds induced by orbital vascular malformation in two cases.

PURPOSE: To investigate the alterations in the retinal pigment epithelium (RPE) in the crest and trough portions of chorioretinal folds (CRFs) induced by an orbital vascular tumor. METHODS: Review of multimodal imaging in two eyes of two patients with globe compression and CRFs from an orbital vascular tumor. RESULTS: Fundus photography demonstrated obliquely extending CRFs with alternating hyperpigmented and hypopigmented linear alterations in both eyes. Fundus autofluorescence (AF) imaging showed obliquely oriented hypoAF lines, incompletely alternating with hyperAF lines. In Case 1, the hyperAF lines had interspersed hypoAF segments and Case 2 had peripapillary mottling of AF. Fluorescein angiography (FA) showed alternating hyper and hypofluorescent lines in the late phase in Case 1. Optical coherence tomography (OCT) documented relative thinning of RPE at the folded crests in Case 1 and preservation of RPE in Case 2. Swept-source OCT angiography (SS-OCTA) demonstrated oblique hyporeflective lines in the outer retina and choriocapillaris layers in Case 2. These findings suggest that the crest of a CRF represents thinned or rarified RPE with hypoAF, transmission hyperfluorescence (FA), partially attenuated RPE layer (OCT), and isoreflectivity (SS-OCTA) while the trough represents compressed RPE with irregular hyperAF, transmission hypofluorescence (FA), thickened RPE layer (OCT), and hyporeflectivity (SS-OCTA). CONCLUSION: The anatomic and functional status of the RPE in CRFs based on multimodal imaging reveals normal to attenuated RPE with hypofunctionality at the fold crest and compacted, thickened RPE at the trough with segmental functional impairment on AF imaging. Anatomic information regarding CRFs is evident on OCT, FA, and SS-OCTA while the functional status is depicted on AF.

CHOROIDAL NEOVASCULAR MEMBRANE AFTER FINE-NEEDLE ASPIRATION BIOPSY OF VITREORETINAL LYMPHOMA.

PURPOSE: To report development of choroidal neovascular membrane at the site of diagnostic transvitreal fine-needle aspiration biopsy of subretinal pigment epithelial infiltrates in an eye with vitreoretinal lymphoma. METHODS: Case report. A 75-year-old white woman with 14-month history of bilateral vitritis. RESULTS: Examination showed vitreous infiltration in both eyes and yellow-white subretinal pigment epithelial infiltrates temporally in the left eye. Visual acuity was 20/400 in both eyes. Transvitreal fine-needle aspiration biopsy of subretinal pigment epithelial infiltrate in the left eye using a long 27-gauge needle attached to a 10-mL syringe revealed large B-cell lymphoma. Treatment was initiated with monthly intravitreal injections of melphalan (10 µg/0.05 mL) to both eyes. One month after first injection, visual acuity in the left eye had decreased to hand motion, and new ill-defined patches of retinal whitening were noticeable in the temporal macular area. Intravenous fluorescein angiography and optical coherence tomography angiography showed large choroidal neovascular membrane temporally at the site of previous fine-needle aspiration biopsy. Four monthly injections of intravitreal bevacizumab (1.25 mg/0.05 mL) resulted in complete regression of choroidal neovascular membrane with improvement of visual acuity to 20/100 in the left eye. CONCLUSION: This case demonstrates the rare development of choroidal neovascular membrane at the site of diagnostic transvitreal fine-needle aspiration biopsy of subretinal pigment epithelial infiltrates in an eye with vitreoretinal lymphoma.

Photodynamic Therapy for Retinal Hemangioblastoma: Treatment Outcomes of 17 Consecutive Patients.

PURPOSE: To report the efficacy of photodynamic therapy (PDT) for management of retinal hemangioblastoma. DESIGN: Retrospective case series. PARTICIPANTS: Seventeen patients with retinal hemangioblastoma treated with PDT. METHODS: The medical records of 17 patients with retinal hemangioblastoma treated with PDT were reviewed, and treatment outcomes were assessed. Photodynamic therapy was performed with 6 mg/m2 body surface area of verteporfin infused intravenously over 10 minutes activated by 50 J/cm2 laser light at 689 nm for 83 or 166 seconds. MAIN OUTCOME MEASURES: Tumor control, subretinal and intraretinal fluid resolution, and visual outcome. RESULTS: Eighteen retinal hemangioblastomas in 17 eyes were treated with PDT. Median patient age was 31 years (mean, 36 years; range, 7-66 years), and median follow-up was 51 months (mean, 61 months; range, 2-144 months). Genetic testing confirmed von Hippel-Lindau disease in 8 of 17 patients (47%). The tumors were unilateral in all patients and unifocal in most patients (n = 13/17 [76%]). The tumor median basal diameter was 3.5 mm (mean, 3.4 mm; range, 1.5-6.0 mm) and median thickness was 2.1 mm (mean, 2.3 mm; range, 1.0-5.0 mm). The tumor location was juxtapapillary in 9 of 18 cases (50%). Associated findings included subretinal fluid (n = 14/17 [82%]) and macular edema (n = 12/17 [71%]). The median number of PDT sessions was 1.5 (mean, 1.8; range, 1.0-4.0). Standard duration of PDT (83 seconds) was used in all cases except 2, in which double duration (166 seconds) was used. Outcomes revealed tumor control in 13 of 18 tumors (72%), partial or complete resolution of subretinal fluid in 10 of 14 eyes (71%), and partial or complete resolution of macular edema in 7 of 12 eyes (58%), and stable or improved visual acuity in 12 of 17 eyes (71%). Photodynamic therapy-related transient exudative response was noted in 4 of 17 eyes (24%). CONCLUSIONS: Photodynamic therapy is an effective treatment for both juxtapapillary and peripheral retinal hemangioblastomas, providing satisfactory rates of tumor control and visual stabilization and improvement. Patients should be monitored for PDT-related transient exudative response.

Conjunctival melanoma: outcomes based on tumour origin in 629 patients at a single ocular oncology centre.

PURPOSE: To investigate clinical features and outcomes of conjunctival melanoma classified by tumour origin. METHODS: Retrospective review of conjunctival melanoma patients at a single ocular oncology centre between April 18, 1974 and September 9, 2019. Lesions were divided into three tumour origin groups (primary acquired melanosis [PAM], nevus, and de novo) and clinical features and outcomes were compared. RESULTS: There were 629 patients with conjunctival melanoma that arose from PAM (n = 476, 76%), nevus (n = 59, 9%), or de novo (n = 94, 15%). A comparison (PAM vs. nevus vs. de novo) revealed patients with tumours arising from PAM presented with older mean age (62 vs. 52 vs. 55 years, p < 0.001), worse initial logMAR visual acuity (Snellen equivalent 20/30 vs. 20/25 vs. 20/25, p = 0.03), and greater clock hour involvement (4.8 vs. 4.0 vs. 3.2, p < 0.001). Tumours arising from nevus had lower frequency of fornix (31% vs. 9% vs. 24%, p = 0.02) and tarsal involvement (29% vs. 9% vs. 26%, p = 0.046) and more frequent classification as AJCC category T1 (60% vs. 89% vs. 62%, p = 0.01). After follow-up of (57.2 vs. 68.2 vs. 51.7 months, p = 0.35), tumours arising from PAM had worse mean final visual acuity (20/50 vs. 20/40 vs. 20/40, p = 0.02) and greater frequency of visual acuity loss ≥3 lines (25% vs. 15% vs. 10%, p = 0.02). Kaplan-Meier estimates for 5-year risk showed no difference by tumour origin for visual acuity loss ≥3 lines, local tumour recurrence, exenteration, metastasis, or death. CONCLUSIONS: Conjunctival melanoma most often arose from PAM, and tumour origin did not affect clinical outcomes.