ABSTRACT
BACKGROUND & AIMS: Large language models including Chat Generative Pretrained Transformers version 4 (ChatGPT4) improve access to artificial intelligence, but their impact on the clinical practice of gastroenterology is undefined. This study compared the accuracy, concordance, and reliability of ChatGPT4 colonoscopy recommendations for colorectal cancer rescreening and surveillance with contemporary guidelines and real-world gastroenterology practice. METHODS: History of present illness, colonoscopy data, and pathology reports from patients undergoing procedures at 2 large academic centers were entered into ChatGPT4 and it was queried for the next recommended colonoscopy follow-up interval. Using the McNemar test and inter-rater reliability, we compared the recommendations made by ChatGPT4 with the actual surveillance interval provided in the endoscopist's procedure report (gastroenterology practice) and the appropriate US Multisociety Task Force (USMSTF) guidance. The latter was generated for each case by an expert panel using the clinical information and guideline documents as reference. RESULTS: Text input of de-identified data into ChatGPT4 from 505 consecutive patients undergoing colonoscopy between January 1 and April 30, 2023, elicited a successful follow-up recommendation in 99.2% of the queries. ChatGPT4 recommendations were in closer agreement with the USMSTF Panel (85.7%) than gastroenterology practice recommendations with the USMSTF Panel (75.4%) (P < .001). Of the 14.3% discordant recommendations between ChatGPT4 and the USMSTF Panel, recommendations were for later screening in 26 (5.1%) and for earlier screening in 44 (8.7%) cases. The inter-rater reliability was good for ChatGPT4 vs USMSTF Panel (Fleiss κ, 0.786; 95% CI, 0.734-0.838; P < .001). CONCLUSIONS: Initial real-world results suggest that ChatGPT4 can define routine colonoscopy screening intervals accurately based on verbatim input of clinical data. Large language models have potential for clinical applications, but further training is needed for broad use.
ABSTRACT
BACKGROUND: The utility of adjuvant chemotherapy (AC) after neoadjuvant therapy and curative intent surgery for clinical stage IIIA (cIIIA) non-small cell lung cancer (NSCLC) is not defined. We sought to evaluate the contribution of AC to overall survival (OS) in patients with cIIIA NSCLC who underwent neoadjuvant therapy followed by curative intent surgical resection. METHODS: The National Cancer Database was queried from 2010 to 2016 for patients with cIIIA NSCLC who underwent curative intent surgical resection after neoadjuvant therapy. Patients were grouped by receipt of AC, and OS was calculated using the Kaplan-Meier method. The association between mortality and AC was evaluated using Cox regression. Ninety-day landmark and propensity score-matched analyses were performed to address bias associated with early postoperative morbidity and mortality. RESULTS: Of 3847 patients who met the inclusion criteria, 780 received AC (20.2%). In the unadjusted cohort there was no difference in 5-year OS between the AC and no AC groups (42.8% vs 43.9%, P = .105). Cox regression demonstrated a decreased risk of mortality in pN > 0 patients receiving AC (hazard ratio, 0.79; 95% confidence interval, 0.68-0.92; P < .003), whereas no difference was seen in node-negative patients (hazard ratio, 0.95; 95% confidence interval, 0.78-1.17; P = .64). In the propensity score-matched groups OS was significantly increased in pN > 0 patients who received AC (5-year OS: 42.4% vs 37%, P < .01), whereas no survival benefit was seen in those who were pN0. CONCLUSIONS: For patients with completely resected cIIIA NSCLC after neoadjuvant therapy, AC is associated with an increase in OS for patients with residual pathologic lymph node involvement.
