Artificial intelligence (AI) tools such as ChatGPT are increasingly applied in digital health and patient education, yet their alignment with established clinical guidelines for cancer-related nutritional management remains unclear. This study aimed to evaluate the concordance, functional characteristics, patient accessibility, and innovation of ChatGPT-generated nutritional recommendations compared with clinical guidelines from the Chinese Society of Clinical Oncology (CSCO), Chinese Nutrition Society (CNS), and European Society for Clinical Nutrition and Metabolism (ESPEN). We analyzed ChatGPT responses across six key nutrition-related issues—anorexia/cachexia, dysphagia, oral mucositis, unintentional weight loss, gastrointestinal intolerance, and nutritional monitoring—and compared them with guideline recommendations. Expert evaluation (n = 5), readability metrics, semantic similarity (TF-IDF), and patient-centered assessments were conducted to compare personalization, innovation, clinical feasibility, evidence-based support, population applicability, clarity, and self-management guidance. ChatGPT recommendations aligned with at least one guideline in 50.0–64.3% of cases, highest for dysphagia (64.3%), and included general strategies such as small frequent meals, texture modification, hydration, and high-protein/high-calorie intake. ChatGPT-specific suggestions (8.3–18.2%) focused on lifestyle and behavioral interventions, including mindful eating, music therapy, and wearable diet trackers. Expert ratings indicated higher personalization (4.3/5) and innovation (4.6/5) for ChatGPT, whereas guidelines scored higher for clinical feasibility (4.7/5), evidence-based support (4.9/5), and population applicability (4.8/5). Between-group differences were statistically significant for clinical feasibility, evidence-based support, and applicability (all p < 0.01; 95% CI for mean differences: 0.62–1.12), whereas personalization showed no significant difference (p = 0.063). ChatGPT exhibited superior patient-centered performance in clarity (4.5 vs. 3.2, p = 0.004, 95% CI: 0.47–2.13) and self-management guidance (4.6 vs. 3.0, p = 0.002, 95% CI: 0.65–2.05) and demonstrated more concise, readable content (Flesch–Kincaid grade 12.9–14.2) compared with guidelines (17.9–20.5). Semantic analysis revealed moderate overlap with CSCO (≈ 0.63) and CNS (≈ 0.59), and lower similarity with ESPEN (≈ 0.47), highlighting ChatGPT’s use of patient-friendly language. Topic modeling identified three clusters: patient support and accessibility (ChatGPT), technical nutrition therapy (ESPEN/CSCO), and nutritional assessment and monitoring (CNS). ChatGPT provides personalized, innovative, and patient-accessible nutritional guidance for cancer-related malnutrition, complementing traditional clinical guidelines. While guidelines remain essential for evidence-based decision-making, AI tools may enhance patient education, engagement, and self-management in digital health applications.

The TGF-β signaling pathway plays a dual role in cancer, acting as both a tumor suppressor and a potent driver of metastasis and immune evasion. Traditional TGF-β inhibitors, primarily ATP-competitive kinase blockers, have faced clinical hurdles including systemic toxicities and compensatory resistance. This review provides a medicinal chemistry-oriented perspective on the fundamental transition from occupancy-based inhibition to event-driven targeted protein degradation (TPD). We systematically analyze the chemical evolution of TGF-β modulators, highlighting how rational SAR optimization—such as linker rigidification and the selection of tissue-specific E3 ligases—enhances ternary complex stability and catalytic efficiency (DC50). Furthermore, we discuss the transformative role of artificial intelligence (AI) and molecular docking in guiding the design of next-generation bifunctional degraders, including PROTACs and LYTACs. By synthesizing cuttingedge research and clinical data, this article outlines a roadmap for dismantling the TGF-β axis, offering strategic insights for the development of more potent, selective, and safer therapeutic agents.

Background Artificial intelligence (AI) tools such as ChatGPT are increasingly applied in digital health and patient education, yet their alignment with established clinical guidelines for cancer-related nutritional management remains unclear. Objective This study aimed to evaluate the concordance, functional characteristics, patient accessibility, and innovation of ChatGPT-generated nutritional recommendations compared with clinical guidelines from the Chinese Society of Clinical Oncology (CSCO), Chinese Nutrition Society (CNS), and European Society for Clinical Nutrition and Metabolism (ESPEN). Methods We analyzed ChatGPT responses across six key nutrition-related issues—anorexia/cachexia, dysphagia, oral mucositis, unintentional weight loss, gastrointestinal intolerance, and nutritional monitoring—and compared them with guideline recommendations. Expert evaluation (n = 5), readability metrics, semantic similarity (TF-IDF), and patient-centered assessments were conducted to compare personalization, innovation, clinical feasibility, evidence-based support, population applicability, clarity, and self-management guidance. Results ChatGPT recommendations aligned with at least one guideline in 50.0–64.3% of cases, highest for dysphagia (64.3%), and included general strategies such as small frequent meals, texture modification, hydration, and high-protein/high-calorie intake. ChatGPT-specific suggestions (8.3–18.2%) focused on lifestyle and behavioral interventions, including mindful eating, music therapy, and wearable diet trackers. Expert ratings indicated higher personalization (4.3/5) and innovation (4.6/5) for ChatGPT, whereas guidelines scored higher for clinical feasibility (4.7/5), evidence-based support (4.9/5), and population applicability (4.8/5). ChatGPT exhibited superior patient-centered performance in clarity (4.5 vs 3.2) and self-management guidance (4.6 vs 3.0) and demonstrated more concise, readable content (Flesch–Kincaid grade 12.9–14.2) compared with guidelines (17.9–20.5). Semantic analysis revealed moderate overlap with CSCO (≈ 0.63) and CNS (≈ 0.59), and lower similarity with ESPEN (≈ 0.47), highlighting ChatGPT’s use of patient-friendly language. Topic modeling identified three clusters: patient support and accessibility (ChatGPT), technical nutrition therapy (ESPEN/CSCO), and nutritional assessment and monitoring (CNS). Conclusions ChatGPT provides personalized, innovative, and patient-accessible nutritional guidance for cancer-related malnutrition, complementing traditional clinical guidelines. While guidelines remain essential for evidence-based decision-making, AI tools may enhance patient education, engagement, and self-management in digital health applications.

BACKGROUND Artificial intelligence (AI)-enabled health decision support systems are increasingly integrated into digital health services, supporting symptom assessment, health information seeking, and medical decision assistance. However, existing technology acceptance models primarily describe adoption as a gradual process driven by perceived usefulness, usability, and performance expectations. These models provide limited explanations for why individuals may readily use AI for routine health information tasks while preferring human oversight when AI systems become involved in higher-risk medical decisions. OBJECTIVE This study aimed to investigate nonlinear trust recalibration patterns in AI-enabled health decision support and examine how perceived decision risk and digital health literacy influence AI reliance and preferences for human oversight. METHODS A nationwide cross-sectional survey was conducted among 2,009 Chinese adults using an online questionnaire platform. Participants evaluated AI-enabled health decision scenarios involving different levels of perceived clinical risk. Associations between decision risk, digital health literacy, trust-related factors, and AI reliance patterns were examined using multivariate regression, correlation analyses, and nonlinear transition analyses. RESULTS The findings demonstrated a nonlinear trust recalibration pattern in AI-enabled health decision support. Participants showed relatively high reliance on AI systems for low-risk information-processing tasks; however, AI reliance decreased substantially when decisions involved greater uncertainty, responsibility, and potential health consequences, representing a “Trust Cliff” phenomenon. Furthermore, digitally literate users demonstrated a dual cognitive pattern: greater AI utilization was accompanied by increased scrutiny regarding algorithmic limitations, transparency, privacy protection, and governance, representing a “Rational Paradox”. CONCLUSIONS AI acceptance in digital health does not follow a simple linear progression. Instead, users continuously recalibrate trust according to decision risk, perceived responsibility, and cognitive evaluation capacity. Sustainable implementation of AI-enabled health systems requires balancing technological capability with appropriate human oversight, transparency, and accountability mechanisms.