ChatGPT outscored human candidates in a virtual objective structured clinical examination in obstetrics and gynecology.

BACKGROUND: Natural language processing is a form of artificial intelligence that allows human users to interface with a machine without using complex codes. The ability of natural language processing systems, such as ChatGPT, to successfully engage with healthcare systems requiring fluid reasoning, specialist data interpretation, and empathetic communication in an unfamiliar and evolving environment is poorly studied. This study investigated whether the ChatGPT interface could engage with and complete a mock objective structured clinical examination simulating assessment for membership of the Royal College of Obstetricians and Gynaecologists. OBJECTIVE: This study aimed to determine whether ChatGPT, without additional training, would achieve a score at least equivalent to that achieved by human candidates who sat for virtual objective structured clinical examinations in Singapore. STUDY DESIGN: This study was conducted in 2 phases. In the first phase, a total of 7 structured discussion questions were selected from 2 historical cohorts (cohorts A and B) of objective structured clinical examination questions. ChatGPT was examined using these questions and responses recorded in a script. Of note, 2 human candidates (acting as anonymizers) were examined on the same questions using videoconferencing, and their responses were transcribed verbatim into written scripts. The 3 sets of response scripts were mixed, and each set was allocated to 1 of 3 human actors. In the second phase, actors were used to presenting these scripts to examiners in response to the same examination questions. These responses were blind scored by 14 qualified examiners. ChatGPT scores were unblinded and compared with historical human candidate performance scores. RESULTS: The average score given to ChatGPT by 14 examiners was 77.2%. The average historical human score (n=26 candidates) was 73.7 %. ChatGPT demonstrated sizable performance improvements over the average human candidate in several subject domains. The median time taken for ChatGPT to complete each station was 2.54 minutes, well before the 10 minutes allowed. CONCLUSION: ChatGPT generated factually accurate and contextually relevant structured discussion answers to complex and evolving clinical questions based on unfamiliar settings within a very short period. ChatGPT outperformed human candidates in several knowledge areas. Not all examiners were able to discern between human and ChatGPT responses. Our data highlight the emergent ability of natural language processing models to demonstrate fluid reasoning in unfamiliar environments and successfully compete with human candidates that have undergone extensive specialist training.

Diaryl piperidines as CB1 receptor antagonists.

The syntheses and SAR investigations of novel CB(1) receptor antagonists based on a 1,2-diaryl piperidine core have been described. Optimization of this core afforded a compound with robust in vivo potency by reducing food intake in a mouse DIO model.

Tetrahydroquinoline sulfonamides as vasopressin 1b receptor antagonists.

Vasopressin 1b (V1b) antagonists have been postulated as possible treatments for depression and anxiety. A novel series of potent and selective V1b antagonists has been identified starting from an in-house screen hit. The incorporation of a sulfonamide linker between a tetrahydroisoquinoline core and amino piperidine lead to the identification of a V1b antagonist with similar affinity for human and rat receptors. Further optimization of the right hand portion afforded potent V1b antagonists that possessed moderate to high selectivity over other receptors.

Discovery of a 3-(4-Pyrimidinyl) Indazole (MLi-2), an Orally Available and Selective Leucine-Rich Repeat Kinase 2 (LRRK2) Inhibitor that Reduces Brain Kinase Activity.

Leucine-rich repeat kinase 2 (LRRK2) is a large, multidomain protein which contains a kinase domain and GTPase domain among other regions. Individuals possessing gain of function mutations in the kinase domain such as the most prevalent G2019S mutation have been associated with an increased risk for the development of Parkinson's disease (PD). Given this genetic validation for inhibition of LRRK2 kinase activity as a potential means of affecting disease progression, our team set out to develop LRRK2 inhibitors to test this hypothesis. A high throughput screen of our compound collection afforded a number of promising indazole leads which were truncated in order to identify a minimum pharmacophore. Further optimization of these indazoles led to the development of MLi-2 (1): a potent, highly selective, orally available, brain-penetrant inhibitor of LRRK2.

Discovery of the 3-Imino-1,2,4-thiadiazinane 1,1-Dioxide Derivative Verubecestat (MK-8931)-A ß-Site Amyloid Precursor Protein Cleaving Enzyme 1 Inhibitor for the Treatment of Alzheimer's Disease.

Verubecestat 3 (MK-8931), a diaryl amide-substituted 3-imino-1,2,4-thiadiazinane 1,1-dioxide derivative, is a high-affinity ß-site amyloid precursor protein cleaving enzyme 1 (BACE1) inhibitor currently undergoing Phase 3 clinical evaluation for the treatment of mild to moderate and prodromal Alzheimer's disease. Although not selective over the closely related aspartyl protease BACE2, verubecestat has high selectivity for BACE1 over other key aspartyl proteases, notably cathepsin D, and profoundly lowers CSF and brain Aß levels in rats and nonhuman primates and CSF Aß levels in humans. In this annotation, we describe the discovery of 3, including design, validation, and selected SAR around the novel iminothiadiazinane dioxide core as well as aspects of its preclinical and Phase 1 clinical characterization.

Discovery of an Orally Available, Brain Penetrant BACE1 Inhibitor that Affords Robust CNS Aß Reduction.

Inhibition of BACE1 to prevent brain Aß peptide formation is a potential disease-modifying approach to the treatment of Alzheimer's disease. Despite over a decade of drug discovery efforts, the identification of brain-penetrant BACE1 inhibitors that substantially lower CNS Aß levels following systemic administration remains challenging. In this report we describe structure-based optimization of a series of brain-penetrant BACE1 inhibitors derived from an iminopyrimidinone scaffold. Application of structure-based design in tandem with control of physicochemical properties culminated in the discovery of compound 16, which potently reduced cortex and CSF Aß40 levels when administered orally to rats.