The BRCA Tumor Suppressor Network in Chromosome Damage Repair by Homologous Recombination.

Mutations in the BRCA1 and BRCA2 genes predispose afflicted individuals to breast, ovarian, and other cancers. The BRCA-encoded products form complexes with other tumor suppressor proteins and with the recombinase enzyme RAD51 to mediate chromosome damage repair by homologous recombination and also to protect stressed DNA replication forks against spurious nucleolytic attrition. Understanding how the BRCA tumor suppressor network executes its biological functions would provide the foundation for developing targeted cancer therapeutics, but progress in this area has been greatly hampered by the challenge of obtaining purified BRCA complexes for mechanistic studies. In this article, we review how recent effort begins to overcome this technical challenge, leading to functional and structural insights into the biochemical attributes of these complexes and the multifaceted roles that they fulfill in genome maintenance. We also highlight the major mechanistic questions that remain.

IDR-targeting compounds suppress HPV genome replication via disruption of phospho-BRD4 association with DNA damage response factors.

Compounds binding to the bromodomains of bromodomain and extra-terminal (BET) family proteins, particularly BRD4, are promising anticancer agents. Nevertheless, side effects and drug resistance pose significant obstacles in BET-based therapeutics development. Using high-throughput screening of a 200,000-compound library, we identified small molecules targeting a phosphorylated intrinsically disordered region (IDR) of BRD4 that inhibit phospho-BRD4 (pBRD4)-dependent human papillomavirus (HPV) genome replication in HPV-containing keratinocytes. Proteomic profiling identified two DNA damage response factors-53BP1 and BARD1-crucial for differentiation-associated HPV genome amplification. pBRD4-mediated recruitment of 53BP1 and BARD1 to the HPV origin of replication occurs in a spatiotemporal and BRD4 long (BRD4-L) and short (BRD4-S) isoform-specific manner. This recruitment is disrupted by phospho-IDR-targeting compounds with little perturbation of the global transcriptome and BRD4 chromatin landscape. The discovery of these protein-protein interaction inhibitors (PPIi) not only demonstrates the feasibility of developing PPIi against phospho-IDRs but also uncovers antiviral agents targeting an epigenetic regulator essential for virus-host interaction and cancer development.

BLM and BRCA1-BARD1 coordinate complementary mechanisms of joint DNA molecule resolution.

The Bloom syndrome helicase BLM interacts with topoisomerase IIIα (TOP3A), RMI1, and RMI2 to form the BTR complex, which dissolves double Holliday junctions and DNA replication intermediates to promote sister chromatid disjunction before cell division. In its absence, structure-specific nucleases like the SMX complex (comprising SLX1-SLX4, MUS81-EME1, and XPF-ERCC1) can cleave joint DNA molecules instead, but cells deficient in both BTR and SMX are not viable. Here, we identify a negative genetic interaction between BLM loss and deficiency in the BRCA1-BARD1 tumor suppressor complex. We show that this is due to a previously overlooked role for BARD1 in recruiting SLX4 to resolve DNA intermediates left unprocessed by BLM in the preceding interphase. Consequently, cells with defective BLM and BRCA1-BARD1 accumulate catastrophic levels of chromosome breakage and micronucleation, leading to cell death. Thus, we reveal mechanistic insights into SLX4 recruitment to DNA lesions, with potential clinical implications for treating BRCA1-deficient tumors.

Mechanisms of RNF168 nucleosome recognition and ubiquitylation.

RNF168 plays a central role in the DNA damage response (DDR) by ubiquitylating histone H2A at K13 and K15. These modifications direct BRCA1-BARD1 and 53BP1 foci formation in chromatin, essential for cell-cycle-dependent DNA double-strand break (DSB) repair pathway selection. The mechanism by which RNF168 catalyzes the targeted accumulation of H2A ubiquitin conjugates to form repair foci around DSBs remains unclear. Here, using cryoelectron microscopy (cryo-EM), nuclear magnetic resonance (NMR) spectroscopy, and functional assays, we provide a molecular description of the reaction cycle and dynamics of RNF168 as it modifies the nucleosome and recognizes its ubiquitylation products. We demonstrate an interaction of a canonical ubiquitin-binding domain within full-length RNF168, which not only engages ubiquitin but also the nucleosome surface, clarifying how such site-specific ubiquitin recognition propels a signal amplification loop. Beyond offering mechanistic insights into a key DDR protein, our study aids in understanding site specificity in both generating and interpreting chromatin ubiquitylation.

Crucial roles of the BRCA1-BARD1 E3 ubiquitin ligase activity in homology-directed DNA repair.

The tumor-suppressor breast cancer 1 (BRCA1) in complex with BRCA1-associated really interesting new gene (RING) domain 1 (BARD1) is a RING-type ubiquitin E3 ligase that modifies nucleosomal histone and other substrates. The importance of BRCA1-BARD1 E3 activity in tumor suppression remains highly controversial, mainly stemming from studying mutant ligase-deficient BRCA1-BARD1 species that we show here still retain significant ligase activity. Using full-length BRCA1-BARD1, we establish robust BRCA1-BARD1-mediated ubiquitylation with specificity, uncover multiple modes of activity modulation, and construct a truly ligase-null variant and a variant specifically impaired in targeting nucleosomal histones. Cells expressing either of these BRCA1-BARD1 separation-of-function alleles are hypersensitive to DNA-damaging agents. Furthermore, we demonstrate that BRCA1-BARD1 ligase is not only required for DNA resection during homology-directed repair (HDR) but also contributes to later stages for HDR completion. Altogether, our findings reveal crucial, previously unrecognized roles of BRCA1-BARD1 ligase activity in genome repair via HDR, settle prior controversies regarding BRCA1-BARD1 ligase functions, and catalyze new efforts to uncover substrates related to tumor suppression.

Structural insight into BRCA1-BARD1 complex recruitment to damaged chromatin.

The BRCA1-BARD1 complex directs the DNA double-strand break (DSB) repair pathway choice to error-free homologous recombination (HR) during the S-G2 stages. Targeting BRCA1-BARD1 to DSB-proximal sites requires BARD1-mediated nucleosome interaction and histone mark recognition. Here, we report the cryo-EM structure of BARD1 bound to a ubiquitinated nucleosome core particle (NCPUb) at 3.1 Å resolution and illustrate how BARD1 simultaneously recognizes the DNA damage-induced mark H2AK15ub and DNA replication-associated mark H4K20me0 on the nucleosome. In vitro and in vivo analyses reveal that the BARD1-NCPUb complex is stabilized by BARD1-nucleosome interaction, BARD1-ubiquitin interaction, and BARD1 ARD domain-BARD1 BRCT domain interaction, and abrogating these interactions is detrimental to HR activity. We further identify multiple disease-causing BARD1 mutations that disrupt BARD1-NCPUb interactions and hence impair HR. Together, this study elucidates the mechanism of BRCA1-BARD1 complex recruitment and retention by DSB-flanking nucleosomes and sheds important light on cancer therapeutic avenues.

Ubiquitin Phosphorylation at Thr12 Modulates the DNA Damage Response.

The ubiquitin system regulates the DNA damage response (DDR) by modifying histone H2A at Lys15 (H2AK15ub) and triggering downstream signaling events. Here, we find that phosphorylation of ubiquitin at Thr12 (pUbT12) controls the DDR by inhibiting the function of 53BP1, a key factor for DNA double-strand break repair by non-homologous end joining (NHEJ). Detectable as a chromatin modification on H2AK15ub, pUbT12 accumulates in nuclear foci and is increased upon DNA damage. Mutating Thr12 prevents the removal of ubiquitin from H2AK15ub by USP51 deubiquitinating enzyme, leading to a pronounced accumulation of ubiquitinated chromatin. Chromatin modified by pUbT12 is inaccessible to 53BP1 but permissive to the homologous recombination (HR) proteins RNF169, RAD51, and the BRCA1/BARD1 complex. Phosphorylation of ubiquitin at Thr12 in the chromatin context is a new histone mark, H2AK15pUbT12, that regulates the DDR by hampering the activity of 53BP1 at damaged chromosomes.

BRCA1/BARD1 intrinsically disordered regions facilitate chromatin recruitment and ubiquitylation.

BRCA1/BARD1 is a tumor suppressor E3 ubiquitin (Ub) ligase with roles in DNA damage repair and in transcriptional regulation. BRCA1/BARD1 RING domains interact with nucleosomes to facilitate mono-ubiquitylation of distinct residues on the C-terminal tail of histone H2A. These enzymatic domains constitute a small fraction of the heterodimer, raising the possibility of functional chromatin interactions involving other regions such as the BARD1 C-terminal domains that bind nucleosomes containing the DNA damage signal H2A K15-Ub and H4 K20me0, or portions of the expansive intrinsically disordered regions found in both subunits. Herein, we reveal novel interactions that support robust H2A ubiquitylation activity mediated through a high-affinity, intrinsically disordered DNA-binding region of BARD1. These interactions support BRCA1/BARD1 recruitment to chromatin and sites of DNA damage in cells and contribute to their survival. We also reveal distinct BRCA1/BARD1 complexes that depend on the presence of H2A K15-Ub, including a complex where a single BARD1 subunit spans adjacent nucleosome units. Our findings identify an extensive network of multivalent BARD1-nucleosome interactions that serve as a platform for BRCA1/BARD1-associated functions on chromatin.

The BRCT Domains of the BRCA1 and BARD1 Tumor Suppressors Differentially Regulate Homology-Directed Repair and Stalled Fork Protection.

The BRCA1 tumor suppressor preserves genome integrity through both homology-directed repair (HDR) and stalled fork protection (SFP). In vivo, BRCA1 exists as a heterodimer with the BARD1 tumor suppressor, and both proteins harbor a phosphate-binding BRCT domain. Here, we compare mice with mutations that ablate BRCT phospho-recognition by Bard1 (Bard1S563F and Bard1K607A) or Brca1 (Brca1S1598F). Brca1S1598F abrogates both HDR and SFP, suggesting that both pathways are likely impaired in most BRCA1 mutant tumors. Although not affecting HDR, the Bard1 mutations ablate poly(ADP-ribose)-dependent recruitment of BRCA1/BARD1 to stalled replication forks, resulting in fork degradation and chromosome instability. Nonetheless, Bard1S563F/S563F and Bard1K607A/K607A mice, unlike Brca1S1598F/S1598F mice, are not tumor prone, indicating that HDR alone is sufficient to suppress tumor formation in the absence of SFP. Nevertheless, because SFP, unlike HDR, is also impaired in heterozygous Brca1/Bard1 mutant cells, SFP and HDR may contribute to distinct stages of tumorigenesis in BRCA1/BARD1 mutation carriers.

BGL3 lncRNA mediates retention of the BRCA1/BARD1 complex at DNA damage sites.

Long non-coding RNAs (lncRNAs) are emerging regulators of genomic stability and human disease. However, the molecular mechanisms by which nuclear lncRNAs directly contribute to DNA damage responses remain largely unknown. Using RNA antisense purification coupled with quantitative mass spectrometry (RAP-qMS), we found that the lncRNA BGL3 binds to PARP1 and BARD1, exhibiting unexpected roles in homologous recombination. Mechanistically, BGL3 is recruited to DNA double-strand breaks (DSBs) by PARP1 at an early time point, which requires its interaction with the DNA-binding domain of PARP1. BGL3 also binds the C-terminal BRCT domain and an internal region (amino acids 127-424) of BARD1, which mediates interaction of the BRCA1/BARD1 complex with its binding partners such as HP1γ and RAD51, resulting in BRCA1/BARD1 retention at DSBs. Cells depleted for BGL3 displayed genomic instability and were sensitive to DNA-damaging reagents. Overall, our findings underscore the biochemical versatility of RNA as a mediator molecule in the DNA damage response pathway, which affects the accumulation of BRCA1/BARD1 at DSBs.

Exploring the role of Large Language Models in haematology: A focused review of applications, benefits and limitations.

Large language models (LLMs) have significantly impacted various fields with their ability to understand and generate human-like text. This study explores the potential benefits and limitations of integrating LLMs, such as ChatGPT, into haematology practices. Utilizing systematic review methodologies, we analysed studies published after 1 December 2022, from databases like PubMed, Web of Science and Scopus, and assessing each for bias with the QUADAS-2 tool. We reviewed 10 studies that applied LLMs in various haematology contexts. These models demonstrated proficiency in specific tasks, such as achieving 76% diagnostic accuracy for haemoglobinopathies. However, the research highlighted inconsistencies in performance and reference accuracy, indicating variability in reliability across different uses. Additionally, the limited scope of these studies and constraints on datasets could potentially limit the generalizability of our findings. The findings suggest that, while LLMs provide notable advantages in enhancing diagnostic processes and educational resources within haematology, their integration into clinical practice requires careful consideration. Before implementing them in haematology, rigorous testing and specific adaptation are essential. This involves validating their accuracy and reliability across different scenarios. Given the field's complexity, it is also critical to continuously monitor these models and adapt them responsively.

Comparison of Large Language Models in Answering Immuno-Oncology Questions: A Cross-Sectional Study.

BACKGROUND: The capability of large language models (LLMs) to understand and generate human-readable text has prompted the investigation of their potential as educational and management tools for patients with cancer and healthcare providers. MATERIALS AND METHODS: We conducted a cross-sectional study aimed at evaluating the ability of ChatGPT-4, ChatGPT-3.5, and Google Bard to answer questions related to 4 domains of immuno-oncology (Mechanisms, Indications, Toxicities, and Prognosis). We generated 60 open-ended questions (15 for each section). Questions were manually submitted to LLMs, and responses were collected on June 30, 2023. Two reviewers evaluated the answers independently. RESULTS: ChatGPT-4 and ChatGPT-3.5 answered all questions, whereas Google Bard answered only 53.3% (P < .0001). The number of questions with reproducible answers was higher for ChatGPT-4 (95%) and ChatGPT3.5 (88.3%) than for Google Bard (50%) (P < .0001). In terms of accuracy, the number of answers deemed fully correct were 75.4%, 58.5%, and 43.8% for ChatGPT-4, ChatGPT-3.5, and Google Bard, respectively (P = .03). Furthermore, the number of responses deemed highly relevant was 71.9%, 77.4%, and 43.8% for ChatGPT-4, ChatGPT-3.5, and Google Bard, respectively (P = .04). Regarding readability, the number of highly readable was higher for ChatGPT-4 and ChatGPT-3.5 (98.1%) and (100%) compared to Google Bard (87.5%) (P = .02). CONCLUSION: ChatGPT-4 and ChatGPT-3.5 are potentially powerful tools in immuno-oncology, whereas Google Bard demonstrated relatively poorer performance. However, the risk of inaccuracy or incompleteness in the responses was evident in all 3 LLMs, highlighting the importance of expert-driven verification of the outputs returned by these technologies.

Functional Analysis of BARD1 Missense Variants on Homology-Directed Repair in Ovarian and Breast Cancers.

Women with germline BRCA1 mutations face an increased risk of developing breast and ovarian cancers. BARD1 (BRCA1 associated RING domain 1) is an essential heterodimeric partner of BRCA1, and mutations in BARD1 are also associated with these cancers. While BARD1 mutations are recognized for their cancer susceptibility, the exact roles of numerous BARD1 missense mutations remain unclear. In this study, we conducted functional assays to assess the homology-directed DNA repair (HDR) activity of all BARD1 missense substitutions identified in 55 breast and ovarian cancer samples, using the real-world data from the COSMIC and cBioPortal databases. Seven BARD1 variants (V85M, P187A, G491R, R565C, P669L, T719R, and Q730L) were confirmed to impair DNA damage repair. Furthermore, cells harboring these BARD1 variants exhibited increased sensitivity to the chemotherapeutic drugs, cisplatin, and olaparib, compared to cells expressing wild-type BARD1. These findings collectively suggest that these seven missense BARD1 variants are likely pathogenic and may respond well to cisplatin-olaparib combination therapy. This study not only enhances our understanding of BARD1's role in DNA damage repair but also offers valuable insights into predicting therapy responses in patients with specific BARD1 missense mutations.

Using artificial intelligence to generate medical literature for urology patients: a comparison of three different large language models.

PURPOSE: Large language models (LLMs) are a form of artificial intelligence (AI) that uses deep learning techniques to understand, summarize and generate content. The potential benefits of LLMs in healthcare is predicted to be immense. The objective of this study was to examine the quality of patient information leaflets (PILs) produced by 3 LLMs on urological topics. METHODS: Prompts were created to generate PILs from 3 LLMs: ChatGPT-4, PaLM 2 (Google Bard) and Llama 2 (Meta) across four urology topics (circumcision, nephrectomy, overactive bladder syndrome, and transurethral resection of the prostate). PILs were evaluated using a quality assessment checklist. PIL readability was assessed by the Average Reading Level Consensus Calculator. RESULTS: PILs generated by PaLM 2 had the highest overall average quality score (3.58), followed by Llama 2 (3.34) and ChatGPT-4 (3.08). PaLM 2 generated PILs were of the highest quality in all topics except TURP and was the only LLM to include images. Medical inaccuracies were present in all generated content including instances of significant error. Readability analysis identified PaLM 2 generated PILs as the simplest (age 14-15 average reading level). Llama 2 PILs were the most difficult (age 16-17 average). CONCLUSION: While LLMs can generate PILs that may help reduce healthcare professional workload, generated content requires clinician input for accuracy and inclusion of health literacy aids, such as images. LLM-generated PILs were above the average reading level for adults, necessitating improvement in LLM algorithms and/or prompt design. How satisfied patients are to LLM-generated PILs remains to be evaluated.

Evaluation of artificial intelligence-generated drug therapy communication skill competencies in medical education.

AIMS: This study compared three artificial intelligence (AI) platforms' potential to identify drug therapy communication competencies expected of a graduating medical doctor. METHODS: We presented three AI platforms, namely, Poe Assistant©, ChatGPT© and Google Bard©, with structured queries to generate communication skill competencies and case scenarios appropriate for graduating medical doctors. These case scenarios comprised 15 prototypical medical conditions that required drug prescriptions. Two authors independently evaluated the AI-enhanced clinical encounters, which integrated a diverse range of information to create patient-centred care plans. Through a consensus-based approach using a checklist, the communication components generated for each scenario were assessed. The instructions and warnings provided for each case scenario were evaluated by referencing the British National Formulary. RESULTS: AI platforms demonstrated overlap in competency domains generated, albeit with variations in wording. The domains of knowledge (basic and clinical pharmacology, prescribing, communication and drug safety) were unanimously recognized by all platforms. A broad consensus among Poe Assistant© and ChatGPT© on drug therapy-related communication issues specific to each case scenario was evident. The consensus primarily encompassed salutation, generic drug prescribed, treatment goals and follow-up schedules. Differences were observed in patient instruction clarity, listed side effects, warnings and patient empowerment. Google Bard did not provide guidance on patient communication issues. CONCLUSIONS: AI platforms recognized competencies with variations in how these were stated. Poe Assistant© and ChatGPT© exhibited alignment of communication issues. However, significant discrepancies were observed in specific skill components, indicating the necessity of human intervention to critically evaluate AI-generated outputs.

Performance of three artificial intelligence (AI)-based large language models in standardized testing; implications for AI-assisted dental education.

INTRODUCTION: The emerging rise in novel computer technologies and automated data analytics has the potential to change the course of dental education. In line with our long-term goal of harnessing the power of AI to augment didactic teaching, the objective of this study was to quantify and compare the accuracy of responses provided by ChatGPT (GPT-4 and GPT-3.5) and Google Gemini, the three primary large language models (LLMs), to human graduate students (control group) to the annual in-service examination questions posed by the American Academy of Periodontology (AAP). METHODS: Under a comparative cross-sectional study design, a corpus of 1312 questions from the annual in-service examination of AAP administered between 2020 and 2023 were presented to the LLMs. Their responses were analyzed using chi-square tests, and the performance was juxtaposed to the scores of periodontal residents from corresponding years, as the human control group. Additionally, two sub-analyses were performed: one on the performance of the LLMs on each section of the exam; and in answering the most difficult questions. RESULTS: ChatGPT-4 (total average: 79.57%) outperformed all human control groups as well as GPT-3.5 and Google Gemini in all exam years (p < .001). This chatbot showed an accuracy range between 78.80% and 80.98% across the various exam years. Gemini consistently recorded superior performance with scores of 70.65% (p = .01), 73.29% (p = .02), 75.73% (p < .01), and 72.18% (p = .0008) for the exams from 2020 to 2023 compared to ChatGPT-3.5, which achieved 62.5%, 68.24%, 69.83%, and 59.27% respectively. Google Gemini (72.86%) surpassed the average scores achieved by first- (63.48% ± 31.67) and second-year residents (66.25% ± 31.61) when all exam years combined. However, it could not surpass that of third-year residents (69.06% ± 30.45). CONCLUSIONS: Within the confines of this analysis, ChatGPT-4 exhibited a robust capability in answering AAP in-service exam questions in terms of accuracy and reliability while Gemini and ChatGPT-3.5 showed a weaker performance. These findings underscore the potential of deploying LLMs as an educational tool in periodontics and oral implantology domains. However, the current limitations of these models such as inability to effectively process image-based inquiries, the propensity for generating inconsistent responses to the same prompts, and achieving high (80% by GPT-4) but not absolute accuracy rates should be considered. An objective comparison of their capability versus their capacity is required to further develop this field of study.

Comparison of the Audiological Knowledge of Three Chatbots: ChatGPT, Bing Chat, and Bard.

INTRODUCTION: The purpose of this study was to evaluate three chatbots - OpenAI ChatGPT, Microsoft Bing Chat (currently Copilot), and Google Bard (currently Gemini) - in terms of their responses to a defined set of audiological questions. METHODS: Each chatbot was presented with the same 10 questions. The authors rated the responses on a Likert scale ranging from 1 to 5. Additional features, such as the number of inaccuracies or errors and the provision of references, were also examined. RESULTS: Most responses given by all three chatbots were rated as satisfactory or better. However, all chatbots generated at least a few errors or inaccuracies. ChatGPT achieved the highest overall score, while Bard was the worst. Bard was also the only chatbot unable to provide a response to one of the questions. ChatGPT was the only chatbot that did not provide information about its sources. CONCLUSIONS: Chatbots are an intriguing tool that can be used to access basic information in a specialized area like audiology. Nevertheless, one needs to be careful, as correct information is not infrequently mixed in with errors that are hard to pick up unless the user is well versed in the field.

Large language models and bariatric surgery patient education: a comparative readability analysis of GPT-3.5, GPT-4, Bard, and online institutional resources.

BACKGROUND: The readability of online bariatric surgery patient education materials (PEMs) often surpasses the recommended 6th grade level. Large language models (LLMs), like ChatGPT and Bard, have the potential to revolutionize PEM delivery. We aimed to evaluate the readability of PEMs produced by U.S. medical institutions compared to LLMs, as well as the ability of LLMs to simplify their responses. METHODS: Responses to frequently asked questions (FAQs) related to bariatric surgery were gathered from top-ranked health institutions. FAQ responses were also generated from GPT-3.5, GPT-4, and Bard. LLMs were then prompted to improve the readability of their initial responses. The readability of institutional responses, initial LLM responses, and simplified LLM responses were graded using validated readability formulas. Accuracy and comprehensiveness of initial and simplified LLM responses were also compared. RESULTS: Responses to 66 FAQs were included. All institutional and initial LLM responses had poor readability, with average reading levels ranging from 9th grade to college graduate. Simplified responses from LLMs had significantly improved readability, with reading levels ranging from 6th grade to college freshman. When comparing simplified LLM responses, GPT-4 responses demonstrated the highest readability, with reading levels ranging from 6th to 9th grade. Accuracy was similar between initial and simplified responses from all LLMs. Comprehensiveness was similar between initial and simplified responses from GPT-3.5 and GPT-4. However, 34.8% of Bard's simplified responses were graded as less comprehensive compared to initial. CONCLUSION: Our study highlights the efficacy of LLMs in enhancing the readability of bariatric surgery PEMs. GPT-4 outperformed other models, generating simplified PEMs from 6th to 9th grade reading levels. Unlike GPT-3.5 and GPT-4, Bard's simplified responses were graded as less comprehensive. We advocate for future studies examining the potential role of LLMs as dynamic and personalized sources of PEMs for diverse patient populations of all literacy levels.

Large language models as assistance for glaucoma surgical cases: a ChatGPT vs. Google Gemini comparison.

PURPOSE: The aim of this study was to define the capability of ChatGPT-4 and Google Gemini in analyzing detailed glaucoma case descriptions and suggesting an accurate surgical plan. METHODS: Retrospective analysis of 60 medical records of surgical glaucoma was divided into "ordinary" (n = 40) and "challenging" (n = 20) scenarios. Case descriptions were entered into ChatGPT and Bard's interfaces with the question "What kind of surgery would you perform?" and repeated three times to analyze the answers' consistency. After collecting the answers, we assessed the level of agreement with the unified opinion of three glaucoma surgeons. Moreover, we graded the quality of the responses with scores from 1 (poor quality) to 5 (excellent quality), according to the Global Quality Score (GQS) and compared the results. RESULTS: ChatGPT surgical choice was consistent with those of glaucoma specialists in 35/60 cases (58%), compared to 19/60 (32%) of Gemini (p = 0.0001). Gemini was not able to complete the task in 16 cases (27%). Trabeculectomy was the most frequent choice for both chatbots (53% and 50% for ChatGPT and Gemini, respectively). In "challenging" cases, ChatGPT agreed with specialists in 9/20 choices (45%), outperforming Google Gemini performances (4/20, 20%). Overall, GQS scores were 3.5 ± 1.2 and 2.1 ± 1.5 for ChatGPT and Gemini (p = 0.002). This difference was even more marked if focusing only on "challenging" cases (1.5 ± 1.4 vs. 3.0 ± 1.5, p = 0.001). CONCLUSION: ChatGPT-4 showed a good analysis performance for glaucoma surgical cases, either ordinary or challenging. On the other side, Google Gemini showed strong limitations in this setting, presenting high rates of unprecise or missed answers.

Generative artificial intelligence chatbots may provide appropriate informational responses to common vascular surgery questions by patients.

OBJECTIVES: Generative artificial intelligence (AI) has emerged as a promising tool to engage with patients. The objective of this study was to assess the quality of AI responses to common patient questions regarding vascular surgery disease processes. METHODS: OpenAI's ChatGPT-3.5 and Google Bard were queried with 24 mock patient questions spanning seven vascular surgery disease domains. Six experienced vascular surgery faculty at a tertiary academic center independently graded AI responses on their accuracy (rated 1-4 from completely inaccurate to completely accurate), completeness (rated 1-4 from totally incomplete to totally complete), and appropriateness (binary). Responses were also evaluated with three readability scales. RESULTS: ChatGPT responses were rated, on average, more accurate than Bard responses (3.08 ± 0.33 vs 2.82 ± 0.40, p < .01). ChatGPT responses were scored, on average, more complete than Bard responses (2.98 ± 0.34 vs 2.62 ± 0.36, p < .01). Most ChatGPT responses (75.0%, n = 18) and almost half of Bard responses (45.8%, n = 11) were unanimously deemed appropriate. Almost one-third of Bard responses (29.2%, n = 7) were deemed inappropriate by at least two reviewers (29.2%), and two Bard responses (8.4%) were considered inappropriate by the majority. The mean Flesch Reading Ease, Flesch-Kincaid Grade Level, and Gunning Fog Index of ChatGPT responses were 29.4 ± 10.8, 14.5 ± 2.2, and 17.7 ± 3.1, respectively, indicating that responses were readable with a post-secondary education. Bard's mean readability scores were 58.9 ± 10.5, 8.2 ± 1.7, and 11.0 ± 2.0, respectively, indicating that responses were readable with a high-school education (p < .0001 for three metrics). ChatGPT's mean response length (332 ± 79 words) was higher than Bard's mean response length (183 ± 53 words, p < .001). There was no difference in the accuracy, completeness, readability, or response length of ChatGPT or Bard between disease domains (p > .05 for all analyses). CONCLUSIONS: AI offers a novel means of educating patients that avoids the inundation of information from "Dr Google" and the time barriers of physician-patient encounters. ChatGPT provides largely valid, though imperfect, responses to myriad patient questions at the expense of readability. While Bard responses are more readable and concise, their quality is poorer. Further research is warranted to better understand failure points for large language models in vascular surgery patient education.