Background: the paucity of early diagnostic markers for clear cell renal cell carcinoma (ccRCC) contributes significantly to its poor clinical prognosis. Aberrations in fatty acid metabolism (FAM) have been implicated in the progression of this malignancy, suggesting the metabolic pathway as a potential target for novel therapeutic approaches. Results: this investigation demonstrated that FAM was decreased in ccRCC, which correlated with a worsening clinical prognosis. A prognostic signature was constructed using FAM-associated long noncoding RNAs (lncRNAs), highlighting notable differences in the tumor microenvironment and treatment responses between high- and low-risk groups. This study employed a suite of six machine learning algorithmsleast absolute shrinkage and selection operator (LASSO) regression, eXtreme gradient boosting (XGBoost), support vector machine recursive feature elimination (SVM-RFE), Random Forest, gradient boosting machine (GBM), and decision treeto identify a significant regulatory axis involving SRD5A3-AS1, hsa-let-7e-5p, and RRM2. The single-cell analysis of ccRCC tissues and peripheral blood, as well as a spatial transcriptomic analysis, revealed that this regulatory axis suppresses FAM in precursor CD8⁺ T cells. These cells exacerbate the clinical outcomes of ccRCC by affecting other immune cells through the macrophage migration inhibitory factor (MIF). Drug sensitivity assays identified axitinib and sorafenib as potential SRD5A3-AS1 inhibitors and dasatinib and fulvestrant as RRM2 inhibitors. The molecular docking results confirmed the stability of the binding of RRM2 to dasatinib and fulvestrant, suggesting that these drugs are promising for therapeutic applications in ccRCC.

Background: Traditional methods for detecting tumor-reactive (TR) CD8 + tumor-infiltrating lymphocytes (TILs) in pancreatic cancer usually focus on neo-antigenic epitopes, which is limited by the narrow range of antigenic epitopes, and the lengthy and complex identification processes, resulting in an incomplete understanding of the biological characteristics of TR CD8 + TILs. Methods: This study introduces a novel approach that integrates single-cell sequencing with deep learning (DL), which enables the identification of tumor-reactive CD8 + T cells without neoantigen screening. The T Cell Receptor Engineered T (TCR-T) cell tumor organoid killing model was employed to validate the functionality of DL-identified TR CD8 + T cells, while spatial transcriptomics was used to confirm receptor-ligand interactions involving TR CD8 + TILs. Results: Comprehensive analyses of TR CD8 + TILs revealed impaired mitochondrial respiratory chain-related pathways regulated by the transcription factor FOS. The TIGIT-NECTIN2 axis was identified as an important immune checkpoint molecule in the tumor microenvironment of pancreatic cancer. T cell receptor (TCR) repertoire analysis demonstrated that some TR CD8 + TILs possess multiple TCR αβ combinations. Furthermore, TCR-T targeting experiments using tumor organoids revealed that combinations of multiple distinct TR TCRs exhibit significantly superior tumor-killing capabilities compared to a single type TCR. Clinically, a higher proportion of TR CD8 + TILs was positively associated with improved responses to neoadjuvant immunotherapy and longer overall survival in pancreatic cancer patients. Conclusion: This study represents a significant advancement in the understanding of TR TIL biology and provides a rapid and accurate method to identify TR CD8 TILs.

Background: Exosomes play a crucial role in intercellular communication in clear cell renal cell carcinoma (ccRCC), while the long non‐coding RNAs (lncRNAs) are implicated in tumorigenesis and progression.Aims: The purpose of this study is to construction a exosomes‐related lncRNA score and a ceRNA network to predict the response to immunotherapy and potential targeted drug in ccRCC.Methods: Data of ccRCC patients were obtained from the TCGA database. Pearson correlation analysis was used to identify eExosomes‐related lncRNAs (ERLRs) from Top10 exosomes‐related genes that have been screened. The entire cohort was randomly divided into a training cohort and a validation cohort in equal scale. LASSO regression and multivariate cox regression was used to construct the ERLRs‐based score. Differences in clinicopathological characteristics, immune microenvironment, immune checkpoints, and drug susceptibility between the high‐ and low‐risk groups were also investigated. Finally, the relevant ceRNA network was constructed by machine learning to analyze their potential targets in immunotherapy and drug use of ccRCC patients.Results: A score consisting of 4ERLRs was identified, and patients with higher ERLRs‐based score tended to have a worse prognosis than those with lower ERLRs‐based score. ROC curves and multivariate Cox regression analysis demonstrated that the score could be considered as a risk factor for prognosis in both training and validation cohorts. Moreover, patients with high scores are predisposed to experience poor overall survival, a larger prevalence of advanced stage (III‐IV), a greater tumor mutational burden, a higher infiltration of immunosuppressive cells, and a greater likelihood of responding favorably to immunotherapy. The importance of EMX2OS was determined by mechanical learning, and the ceRNA network was constructed, and EMX2OS may be a potential therapeutic target, possibly exerting its function through the EMX2OS/hsa‐miR‐31‐5p/TLN2 axis.Conclusions: Based on machine learning, a novel ERLRs‐based score was constructed for predicting the survival of ccRCC patients. The ERLRs‐based score is a promising potential independent prognostic factor that is closely correlated with the immune microenvironment and clinicopathological characteristics. Meanwhile, we screened out key lncRNAEMX2OS and identified the EMX2OS/hsa‐miR‐31‐5p/TLN2 axis, which may provide new clues for the targeted therapy of ccRCC.

Recently, immunogenic cell death (ICD) has been identified as a regulatory cell death mechanism that induces an adaptive immune response, thereby improving enhancing the efficacy of immunotherapy and contributing to improved prognosis in bladder cancer (BLCA). This study established a risk signature based on ICD and identified ICD-related genes as diagnostic markers and therapeutic targets for BLCA. Thirty-two key ICD-risk genes (IRGs) were screened from correlation and univariate Cox regression analyses. Data obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases regarding BLCA and normal samples were categorized based on IRGs. ICD-based prognostic signature was built based on IRGs, stratifying BLCA patients into high- and low-risk groups. SLC2A3 was identified as a hub gene by 101 combinatorial machine learning algorithms and 10 individual machine learning algorithms. Based on single-cell sequencing data, we identified the cancer-associated fibroblasts (CAFs), the key cell population exhibiting high SLC2A3 expression. Functional analyses were performed to explore the potential value of SLC2A3 as a target for precision therapy. A prognostic signature was constructed using IRGs, indicating significant differences in the tumor microenvironment (TME) and treatment response between different risk groups. We identified SLC2A3 as the most critical IRG exhibiting high expression in the fibroblast population of patients with BLCA, especially in CAFs, which play an important role in BLCA progression. We found that inhibiting SLC2A3 expression may enhance the effectiveness of immunotherapy and the identified potential drugs targeting SLC2A3. We demonstrated that the identified IRGs serve as risk factors for clinical prognosis in BLCA and successfully constructed an ICD-based prognostic signature. Additionally, SLC2A3 holds potential as a therapeutic target to advance precision and personalized treatment strategies for BLCA, in combination with immunotherapy.

Traditional face-to-face mental health treatments are often limited by time and space. Thanks to the development of advanced large language models (LLMs), digital mental health treatments can provide personalized advice to patients and improve compliance. However, in the field of CBT-I, specialized, real-time interactive dialogue platforms have not been fully developed. Our research team construct an eCBT-I intelligent dialogue system based on the RAG architecture, aiming to provide an example of the deep integration of CBT-I knowledge graphs and large language models. Furthermore, in order to optimize the performance of the system’s core language generation module on the insomnia dialogue dataset, we systematically include eight mainstream large language models (ChatGLM2-6b, ChatGLM3-6b, Baichuan-7b, Baichuan-13b, Qwen-7b, Qwen2-7b, Llama-2-7b-chat-hf, and Llama-2-13b-chat-hf) and three adaptation strategies (LoRA, QLoRA, and Freeze). We screen the suitability of the three adaptation strategies for the eight major language models in the group, and thus determine the best adaptation method for each language model to maximize performance improvement. The eight best-adapted language models are then evaluated in three dimensions to compare their performance on the small sample sleep dialogue dataset and the C-eval dataset. All subjects that evaluated under experimental conditions are historical medical records and patients who did not exhibit delirium and had normal language expression abilities. Through the matching of model characteristics to adaptation strategies and the horizontal evaluation of multiple models, we compare the contribution of different fine-tuning strategies to the performance improvement of different language models on the small insomnia dialogue dataset, and finally determine that Qwen2-7b (Freeze) is the model with the best performance on the insomnia dialogue dataset. This study effectively integrates the CBT-I knowledge graph with the large language model through the RAG architecture, which improves the professionalism of the eCBT-I intelligent dialogue system. The systematic fine-tuning method selection process and the confirmation of the optimal model not only improve the adaptability of the large language model in the CBT-I task, but also provide a useful paradigm for AI applications in medical subfields with resource constraints and difficulties in data collection, laying a solid foundation for more accurate and efficient digital CBT-I clinical practice in the future.