Evaluating Large Language Models on their Accuracy and Completeness: Immune Checkpoint Inhibitors and their Ocular Toxicities.

PURPOSE: To analyze the accuracy and thoroughness of 3 large language models (LLMs) to produce information for providers about immune checkpoint inhibitor (ICI) ocular toxicities. METHODS: Eight questions were created about the general definition of checkpoint inhibitors, their mechanism of action, ocular toxicities, and toxicity management. All were inputted into ChatGPT 4.0, Bard, and LLaMA programs. Utilizing the 6-point Likert scale for accuracy and completeness, four ophthalmologists who routinely treat ocular toxicities of immunotherapy agents rated the LLMs answers. ANOVA testing was used to assess significant differences among the three LLMs and a post-hoc pairwise t-test. Fleiss kappa values were calculated to account for interrater variability. RESULTS: ChatGPT responses were rated with an average of 4.59 for accuracy and 4.09 for completeness; Bard answers were rated 4.59 and 4.19; LLaMA results were rated 4.38 and 4.03. The three LLMs did not significantly differ in accuracy (p=0.47) nor completeness (p=0.86). Fleiss kappa values were found to be poor for both accuracy (-0.03) and completeness (0.01). CONCLUSIONS: All three LLMs provided highly accurate and complete responses to questions centered on ICI inhibitor ocular toxicities and management. Further studies are needed to assess specific ICI agents and the accuracy and completeness of updated versions of LLMs.

Accuracy and Completeness of Large Language Models About Antibody-Drug Conjugates and Associated Ocular Adverse Effects.

PURPOSE: The purpose of this study was to assess the accuracy and completeness of 3 large language models (LLMs) to generate information about antibody-drug conjugate (ADC)-associated ocular toxicities. METHODS: There were 22 questions about ADCs, tisotumab vedotin, and mirvetuximab soravtansine that were developed and input into ChatGPT 4.0, Bard, and LLaMa. Answers were rated by 4 ocular toxicity experts using standardized 6-point Likert scales on accuracy and completeness. ANOVA tests were conducted for comparison between the 3 subgroups, followed by pairwise t-tests. Interrater variability was assessed with Fleiss kappa tests. RESULTS: The mean accuracy score was 4.62 (SD 0.89) for ChatGPT, 4.77 (SD 0.90) for Bard, and 4.41 (SD 1.09) for LLaMA. Both ChatGPT (P = 0.03) and Bard (P = 0.003) scored significantly better for accuracy when compared with LLaMA. The mean completeness score was 4.43 (SD 0.91) for ChatGPT, 4.57 (SD 0.93) for Bard, and 4.42 (SD 0.99) for LLaMA. There were no significant differences in completeness scores between groups. Fleiss kappa assessment for interrater variability was good (0.74) for accuracy and fair (0.31) for completeness. CONCLUSIONS: All 3 LLMs had relatively high accuracy and completeness ratings, showing LLMs are able to provide sufficient answers for niche topics of ophthalmology. Our results indicate that ChatGPT and Bard may be slightly better at providing more accurate answers than LLaMA. As further research and treatment plans are developed for ADC-associated ocular toxicities, these LLMs should be reassessed to see if they provide complete and accurate answers that remain in line with current medical knowledge.

Ocular toxicities associated with antibody drug conjugates.

PURPOSE OF REVIEW: To review the structure, mechanism of action, and pathophysiology of antibody-drug conjugates (ADCs) used to treat gynecological malignancies associated with ocular adverse effects. RECENT FINDINGS: Recent research shows tisotumab vedotin causes ocular toxicity localized to the conjunctiva, with common adverse effects being conjunctivitis, dry eye, blepharitis, and keratitis. Toxicity is likely due to targeting tissue factor (TF) in the conjunctiva, leading to direct delivery of the cytotoxic payload resulting in apoptosis and bystander killing. Mirvetuximab soravtansine causes blurred vision, keratitis, or dry eye with toxicity often localized in the cornea. Off-target inflammation appears to cause ocular adverse effects, with nonreceptor mediated macropinocytosis by corneal stem cells. SUMMARY: Collaboration between oncologists and ophthalmologists with adherence to mitigation protocols can decrease the risk of ocular adverse events.

Factors influencing treatment and time spent with physicians in patients with uveitis compared to other ophthalmology subspecialties in the National Ambulatory Medical Care Survey.

BACKGROUND: Cases of uveitis can necessitate long-term treatment resulting in recurrent follow-up appointments. Analysing the demographic distribution and patient factors influencing treatment and time spent with physicians in this population compared to other subspecialties of ophthalmology using the National Ambulatory Medical Care Survey (NAMCS) has not previously been studied. METHODS: Data were extracted from the NAMCS database, a large, nationally representative survey of office-based specialists, entered between 2012-2016 and 2018. Demographics, time with physician, and payor types were compared between patients with a uveitis-related diagnosis codes versus all other ophthalmic subspecialty diagnoses. RESULTS: Overall, 12,870 ophthalmic patients were included of which 300 had uveitis-related diagnosis codes. Uveitis patients were more likely to be non-Caucasian (p < 0.0001 to p = 0.022), visiting the physician's office due to flare of or treatment for a chronic medical problem (p < 0.0001 to p = 0.022). Adjusted for age, sex, race, and ethnicity, uveitis patients spent a significantly longer time (mean 27.5 min) compared to comprehensive ophthalmology patients (mean 25.5 min) with their physician (p = 0.0041). Among the uveitis patient population, African American patients (p = 0.0053), Hispanic or Latino (p = 0.034), and Medicaid (p = 0.035) patients had increased office visit times. CONCLUSIONS: Those with uveitis spent more time with the physician than comprehensive patients. Race, ethnicity, payor type, and the major reason for the visit all significantly impacted uveitis office visit times. In order to manage their schedules, providers should be aware of the additional support and time needed by these patients during office visits.

Punctal Atresia As a Clinical Indicator of Systemic Genetic Anomalies.

BACKGROUND: Punctal atresia or agenesis (PA) is a rare congenital anomaly characterized by the absence or closure of the tear duct puncta, potentially linked to systemic genetic anomalies. The necessity of a genetic workup based solely on the presence of PA remains uncertain. This study investigates a cohort of PA patients, examining the prevalence and types of associated syndromes. METHODS: A retrospective medical records review of all patients diagnosed with PA at the Children's Hospital of Philadelphia between 2009-2023 was conducted, analyzing medical histories and genetic testing results. Primary outcomes included the prevalence of systemic syndromes, while secondary outcomes focused on the variety of associated syndromes. RESULTS: Forty-four patients were included, of which 31 were male (70%) with a mean ± SD age 3.3 ± 3.3 years. Overall, 87 puncta in the study cohort were affected, and 26 cases (59%) were bilateral. Systemic abnormalities or genetic syndromes were identified in 19 patients (43%), with the most common being Ectodermal Dysplasia and Down syndrome. Additional rare syndromes were demonstrated. No significant association was found between systemic abnormalities and gender, bilaterality, or the number of puncta involved. CONCLUSIONS: A high incidence of systemic syndromes (43%) was observed in the study cohort. In individuals with PA who also exhibit extraocular disease, systemic evaluation and genetic workup should be considered. Syndromic diagnoses identified in our cohort also include: Branchio-oto-renal syndrome, 22q11.2 deletion syndrome, 1q21.1 microdeletion syndrome, NF1, monosomy 4q and trisomy 6q, which represent novel associations. The lack of correlation between PA's phenotypic severity and systemic abnormalities highlights the need to obtain a comprehensive medical history and consider a systemic workup in PA patients.

Investigating the Accuracy and Completeness of an Artificial Intelligence Large Language Model About Uveitis: An Evaluation of ChatGPT.

PURPOSE: To assess the accuracy and completeness of ChatGPT-generated answers regarding uveitis description, prevention, treatment, and prognosis. METHODS: Thirty-two uveitis-related questions were generated by a uveitis specialist and inputted into ChatGPT 3.5. Answers were compiled into a survey and were reviewed by five uveitis specialists using standardized Likert scales of accuracy and completeness. RESULTS: In total, the median accuracy score for all the uveitis questions (n = 32) was 4.00 (between "more correct than incorrect" and "nearly all correct"), and the median completeness score was 2.00 ("adequate, addresses all aspects of the question and provides the minimum amount of information required to be considered complete"). The interrater variability assessment had a total kappa value of 0.0278 for accuracy and 0.0847 for completeness. CONCLUSION: ChatGPT can provide relatively high accuracy responses for various questions related to uveitis; however, the answers it provides are incomplete, with some inaccuracies. Its utility in providing medical information requires further validation and development prior to serving as a source of uveitis information for patients.

Oxidative stress induces lysosomal membrane permeabilization and ceramide accumulation in retinal pigment epithelial cells.

Oxidative stress has been implicated in the pathogenesis of age-related macular degeneration, the leading cause of blindness in older adults, with retinal pigment epithelium (RPE) cells playing a key role. To better understand the cytotoxic mechanisms underlying oxidative stress, we used cell culture and mouse models of iron overload, as iron can catalyze reactive oxygen species formation in the RPE. Iron-loading of cultured induced pluripotent stem cell-derived RPE cells increased lysosomal abundance, impaired proteolysis and reduced the activity of a subset of lysosomal enzymes, including lysosomal acid lipase (LIPA) and acid sphingomyelinase (SMPD1). In a liver-specific Hepc (Hamp) knockout murine model of systemic iron overload, RPE cells accumulated lipid peroxidation adducts and lysosomes, developed progressive hypertrophy and underwent cell death. Proteomic and lipidomic analyses revealed accumulation of lysosomal proteins, ceramide biosynthetic enzymes and ceramides. The proteolytic enzyme cathepsin D (CTSD) had impaired maturation. A large proportion of lysosomes were galectin-3 (Lgals3) positive, suggesting cytotoxic lysosomal membrane permeabilization. Collectively, these results demonstrate that iron overload induces lysosomal accumulation and impairs lysosomal function, likely due to iron-induced lipid peroxides that can inhibit lysosomal enzymes.

Automated 3D segmentation of guard cells enables volumetric analysis of stomatal biomechanics.

Automating the three-dimensional (3D) segmentation of stomatal guard cells and other confocal microscopy data is extremely challenging due to hardware limitations, hard-to-localize regions, and limited optical resolution. We present a memory-efficient, attention-based, one-stage segmentation neural network for 3D images of stomatal guard cells. Our model is trained end to end and achieved expert-level accuracy while leveraging only eight human-labeled volume images. As a proof of concept, we applied our model to 3D confocal data from a cell ablation experiment that tests the "polar stiffening" model of stomatal biomechanics. The resulting data allow us to refine this polar stiffening model. This work presents a comprehensive, automated, computer-based volumetric analysis of fluorescent guard cell images. We anticipate that our model will allow biologists to rapidly test cell mechanics and dynamics and help them identify plants that more efficiently use water, a major limiting factor in global agricultural production and an area of critical concern during climate change.