Enhancing Early Detection of Cognitive Decline in the Elderly through Ensemble of NLP Techniques: A Comparative Study Utilizing Large Language Models in Clinical Notes.

Background: Early detection of cognitive decline in elderly individuals facilitates clinical trial enrollment and timely medical interventions. This study aims to apply, evaluate, and compare advanced natural language processing techniques for identifying signs of cognitive decline in clinical notes. Methods: This study, conducted at Mass General Brigham (MGB), Boston, MA, included clinical notes from the 4 years prior to initial mild cognitive impairment (MCI) diagnosis in 2019 for patients ≥ 50 years. Note sections regarding cognitive decline were labeled manually. A random sample of 4,949 note sections filtered with cognitive functions-related keywords were used for traditional AI model development, and 200 random subset were used for LLM and prompt development; another random sample of 1996 note sections without keyword filtering were used for testing. Prompt templates for large language models (LLM), Llama 2 on Amazon Web Service and GPT-4 on Microsoft Azure, were developed with multiple prompting approaches to select the optimal LLM-based method. Baseline comparisons were made with XGBoost and a hierarchical attention-based deep neural network model. An ensemble of the three models was then constructed using majority vote. Results: GPT-4 demonstrated superior accuracy and efficiency to Llama 2. The ensemble model outperformed individual models, achieving a precision of 90.3%, recall of 94.2%, and F1-score of 92.2%. Notably, the ensemble model demonstrated a marked improvement in precision (from a 70%-79% range to above 90%) compared to the best performing single model. Error analysis revealed 63 samples were wrongly predicted by at least one model; however, only 2 cases (3.2%) were mutual errors across all models, indicating diverse error profiles among them. Conclusion: Our findings indicate that LLMs and traditional models exhibit diverse error profiles. The ensemble of LLMs and locally trained machine learning models on EHR data was found to be complementary, enhancing performance and improving diagnostic accuracy.

A study of high dose furmonertinib in EGFR exon 20 insertion mutation-positive advanced non-small cell lung cancer.

Background: The epidermal growth factor receptor (EGFR) ex20ins mutation, as a rare subtype of mutation, has gradually attracted attention. Its heterogeneity is high, its prognosis is extremely poor, and the efficacy of existing traditional treatment plans is limited. In this study, we aimed to evaluate efficacy of high dose furmonertinib as a first-line treatment for EGFR ex20ins-positive NSCLC. Methods: This is a retrospective, multi-center, non-interventional study. From May 2021 to March 2023, 9 NSCLC patients with EGFR ex20ins were enrolled. Efficacy and safety of 160 mg furmonertinib were evaluated. Objective response rate (ORR), disease control rate (DCR), median progression-free survival (PFS) and treatment related adverse events (TRAEs) were assessed. Results: Of the evaluated patients, six patients experienced partial remission (PR), two patients experienced stable disease (SD) and one patient experienced progress disease (PD). Data indicated 66.7% ORR and 88.9% DCR. The median progression free survival (PFS) was 7.2 months (95% CI: 6.616 - 7.784). Besides, a longgest PFS with 18 months was found in one patient with p.H773_V774insGTNPH mutation. No ≥ level 3 adverse events have been found. Conclusions: The study proved the potential efficacy of 160mg furmonertinib in patients with advanced NSCLC with EGFR ex20ins. Meanwhile, 160mg furmonertinib had a good safety profile.

Colorimetric and fluorescent independent dual "signal on" biosensor for accurate detection of ochratoxin A based on aptamer-triggered biocatalytic reactions.

Ochratoxin A (OTA) is a hazardous food contaminant with significant health risks. Dual-channel OTA detection is noted for its cross-reference capability and high accuracy. Still, challenges in addressing in-system corrections and "signal off" related false positives and limited signal gains remain. Herein, we developed a dual-channel "signal on" aptasensor with one recognition process and two independent signal outputs for OTA analysis. The OTA aptamer binds to magnetic beads (MBs) and partially hybridizes with a complementary-trigger (cDNA-Trigger) sequence. Adding OTA disrupts the duplex sequence, leading to G-quadruplex (G4) formation and enrichment on the MBs, which then interacts with hemin to catalyze a color signal. Concurrently, the freed cDNA-Trigger catalyzes an enzyme-free DNA circuit, producing a fluorescence signal. The magnetic enrichment and signal amplification strategies make the proposed assay demonstrate excellent sensitivity toward OTA, with limits of detection (LOD) of 0.017 pM in the fluorescence channel and 48.1 pM in the colorimetric channel. Both channels have effectively detected OTA in grape juice and baijiu, demonstrating their applicability and reliability. Moreover, given the widespread use of smartphones globally, a mini-program with a self-correction function was designed to facilitate on-site colorimetric channel monitoring, making OTA detection more accessible and user-friendly.

Re-emergent Tremor in Parkinson's Disease: Evidence of Pathologic ß and Prokinetic γ Activity.

BACKGROUND: Re-emergent tremor is characterized as a continuation of resting tremor and is often highly therapy refractory. This study examines variations in brain activity and oscillatory responses between resting and re-emergent tremors in Parkinson's disease. METHODS: Forty patients with Parkinson's disease (25 males, mean age, 66.78 ± 5.03 years) and 40 age- and sex-matched healthy controls were included in the study. Electroencephalogram and electromyography signals were simultaneously recorded during resting and re-emergent tremors in levodopa on and off states for patients and mimicked by healthy controls. Brain activity was localized using the beamforming technique, and information flow between sources was estimated using effective connectivity. Cross-frequency coupling was used to assess neuronal oscillations between tremor frequency and canonical frequency oscillations. RESULTS: During levodopa on, differences in brain activity were observed in the premotor cortex and cerebellum in both the patient and control groups. However, Parkinson's disease patients also exhibited additional activity in the primary sensorimotor cortex. On withdrawal of levodopa, different source patterns were observed in the supplementary motor area and basal ganglia area. Additionally, levodopa was found to suppress the strength of connectivity (P < 0.001) between the identified sources and influence the tremor frequency-related coupling, leading to a decrease in ß (P < 0.001) and an increase in γ frequency coupling (P < 0.001). CONCLUSIONS: Distinct variations in cortical-subcortical brain activity are evident in tremor phenotypes. The primary sensorimotor cortex plays a crucial role in the generation of re-emergent tremor. Moreover, oscillatory neuronal responses in pathological ß and prokinetic γ activity are specific to tremor phenotypes. © 2024 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

A control study on pain characteristics and influencing factors in patients with depressive disorders-based on a 5-year follow-up report from the epidemiological survey of mental disorders in Shandong Province, China.

BACKGROUND: This study extends from the 2015 Shandong Province Epidemiological Survey of Mental Disorders in adults aged 18 and above. Over five years, it investigates pain characteristics and influencing factors in individuals with depressive disorders in Shandong Province. METHODS: The study encompasses 871 individuals who met DSM-IV criteria for depressive disorders in 2015. Using 1:1:1 matching by gender, age, and residence, 825 non-afflicted individuals were selected as high-risk controls, and 825 screening-negative individuals became low-risk controls. A follow-up study in 2020 involved 1848 participants. Survey tools included a general information questionnaire, General Health Questionnaire-12 (GHQ-12), SCID-I/P, Global Pain Scale (GPS), Quality of Life Questionnaire (QLQ), PSQI, MoCA, and clinical data questionnaire. RESULTS: GPS scores in the current depressive group were higher than in non-current depressive group (Z = 14.36, P < 0.01). GPS scores in study group exceeded those in high-risk and low-risk control groups (H = 93.71, P < 0.01). GPS scores in non-remission group were higher than in the remission group (Z = 8.90, P < 0.01). Regression analysis revealed positive correlations between GPS scores and physical illnesses, current depression, incumbency, GHQ-12 total score, and PSQI total score. Negative correlations were observed with QLQ total score and MoCA total score. LIMITATIONS: The study could not assess pain during the 2015 survey, limiting controlled pain analysis before and after five years. CONCLUSION: Depression sufferers may experience prolonged heightened pain, potentially relieved when depression subsides. Individual pain is influenced by depression, physical illnesses, sleep quality, quality of life, cognitive function, gender, residence, and occupation.

QSAR analysis of VEGFR-2 inhibitors based on machine learning, Topomer CoMFA and molecule docking.

VEGFR-2 kinase inhibitors are clinically approved drugs that can effectively target cancer angiogenesis. However, such inhibitors have adverse effects such as skin toxicity, gastrointestinal reactions and hepatic impairment. In this study, machine learning and Topomer CoMFA, which is an alignment-dependent, descriptor-based method, were employed to build structural activity relationship models of potentially new VEGFR-2 inhibitors. The prediction ac-curacy of the training and test sets of the 2D-SAR model were 82.4 and 80.1%, respectively, with KNN. Topomer CoMFA approach was then used for 3D-QSAR modeling of VEGFR-2 inhibitors. The coefficient of q2 for cross-validation of the model 1 was greater than 0.5, suggesting that a stable drug activity-prediction model was obtained. Molecular docking was further performed to simulate the interactions between the five most promising compounds and VEGFR-2 target protein and the Total Scores were all greater than 6, indicating that they had a strong hydrogen bond interactions were present. This study successfully used machine learning to obtain five potentially novel VEGFR-2 inhibitors to increase our arsenal of drugs to combat cancer.

Advances in phage-host interaction prediction: in silico method enhances the development of phage therapies.

Phages can specifically recognize and kill bacteria, which lead to important application value of bacteriophage in bacterial identification and typing, livestock aquaculture and treatment of human bacterial infection. Considering the variety of human-infected bacteria and the continuous discovery of numerous pathogenic bacteria, screening suitable therapeutic phages that are capable of infecting pathogens from massive phage databases has been a principal step in phage therapy design. Experimental methods to identify phage-host interaction (PHI) are time-consuming and expensive; high-throughput computational method to predict PHI is therefore a potential substitute. Here, we systemically review bioinformatic methods for predicting PHI, introduce reference databases and in silico models applied in these methods and highlight the strengths and challenges of current tools. Finally, we discuss the application scope and future research direction of computational prediction methods, which contribute to the performance improvement of prediction models and the development of personalized phage therapy.

Comparison of the cachexia index based on hand-grip strength (H-CXI) with the original CXI for the prediction of cancer cachexia and prognosis in patients who underwent radical colectomy for colorectal cancer.

Background and aims: The cachexia index (CXI) is a novel biomarker for estimating cancer cachexia. The cachexia index based on hand-grip strength (H-CXI) has been recently developed as a simple proxy for CXI. The present study aims to compare both the H-CXI and CXI for the prediction of cancer cachexia and postoperative outcomes in patients who underwent radical colectomy for colorectal cancer. Methods: Patients who underwent radical operations for colorectal cancer were included in this study. Cancer cachexia was diagnosed according to the international consensus outlined by Fearon et al. The cachexia index (CXI) was calculated as [skeletal muscle index (SMI) × serum albumin/neutrophil-to-lymphocyte ratio (NLR)]. The H-CXI was calculated as [hand-grip strength (HGS)/height2 × serum albumin/NLR]. The SMI was measured based on the preoperative CT images at the third lumbar vertebra (L3) level. HGS was measured before surgery. Results: From July 2014 to May 2021, a total of 1,411 patients were included in the present study, of whom 361 (25.6%) were identified as having cancer cachexia. Patients with cachexia had a lower CXI (p < 0.001) and lower H-CXI (p < 0.001) than those without cachexia. A low CXI but not low H-CXI independently predicted cancer cachexia in the multivariate analysis (OR 1.448, p = 0.024). Both a low CXI (HR 1.476, p < 0.001 for OS; HR 1.611, p < 0.001 for DFS) and low H-CXI (HR 1.369, p = 0.007 for OS; HR 1.642, p < 0.001 for DFS) were independent predictors for overall survival (OS) and disease-free survival (DFS) after adjusting for the same covariates. A low H-CXI but not low CXI was an independent risk factor for postoperative complications (OR 1.337, p = 0.044). No significant association was found between cancer cachexia and postoperative complications. Conclusion: The CXI and H-CXI exhibited better prognostic value than cancer cachexia for the prediction of postoperative outcomes in patients who underwent radical colectomy for colorectal cancer. The H-CXI was a superior index over the CXI in predicting short-term clinical outcomes, whereas the CXI demonstrated a closer correlation with Fearon's criteria for cancer cachexia. Ideal tools for the assessment of cancer cachexia should incorporate not only weight loss but also muscle mass, physical function, and inflammatory state.

Body constitutions of traditional Chinese medicine caused a significant effect on irritable bowel syndrome.

BACKGROUND: According to the theory of traditional Chinese medicine (TCM), all types of body constitutions, except for the Gentleness (i.e., the control group in our study), have disease susceptibility and affect the disease development process. This study attempted to investigate the relationship between TCM body constitutions and irritable bowel syndrome (IBS). METHODS: This cross-sectional study was based on Taiwan Biobank (TWB) and collected clinical data from 13,941 subjects aged 30-70. The results of the study showed that subjects with Yang-deficiency (N=3,161 subjects, OR=2.654, 95% CI=1.740-3.910), Ying-deficiency (N=3,331 subjects, OR=1.096, 95% CI=0.627-1.782) or Stasis (N=2,335 subjects, OR=1.680, 95% CI=0.654-3.520) were more likely to have IBS. RESULTS: If the subjects with two or more TCM body constitutions: Yang-deficiency + Ying-deficiency (OR=3.948, 95% CI=2.742-5.560), Yang-deficiency + Stasis (OR=2.312, 95% CI=1.170-4.112), Ying-deficiency + Stasis (OR=1.851, 95% CI=0.828-3.567), or Yang-deficiency + Ying-deficiency + Stasis (OR=3.826, 95% CI= 2.954-4.932) were also prone to IBS. CONCLUSION: These results confirmed the high correlation between TCM body constitutions and IBS. Because the current treatment for IBS is not entirely satisfactory, integrated traditional Chinese and Western medicine might provide patients with an alternative treatment option to alleviate IBS.

Phase I pharmacokinetic, safety, and preliminary efficacy study of tiragolumab in combination with atezolizumab in Chinese patients with advanced solid tumors.

PURPOSE: Tiragolumab is an immunoglobulin G1 monoclonal antibody targeting the immune checkpoint T cell immunoreceptor with immunoglobulin and immunoreceptor ITIM domains. Targeting multiple immune pathways may improve anti-tumor responses. The phase I YP42514 study assessed the pharmacokinetics (PK), safety, and preliminary efficacy of tiragolumab plus atezolizumab in Chinese patients with advanced solid tumors. METHODS: Adult patients from mainland China with Eastern Cooperative Oncology Group performance score 0/1, life expectancy of ≥ 12 weeks, and adequate hematologic/end organ function were eligible. Patients received tiragolumab 600 mg and atezolizumab 1200 mg intravenous every 3 weeks. Key endpoints were PK (serum concentrations of tiragolumab and atezolizumab) and safety. Results from this study were compared with the global phase I study, GO30103 (NCT02794571). RESULTS: In this study, 20 patients received a median of five doses of tiragolumab plus atezolizumab. Median age was 57.5 years, 85.0% of patients were male and the most common tumor type was non-small cell lung cancer. Exposures in Chinese patients were comparable to the global GO30103 population: geometric mean ratio was 1.07 for Cycle 1 tiragolumab area under the concentration-time curve0-21 and 0.92 and 0.93 for Cycle 1 peak and trough atezolizumab exposure, respectively. Treatment-related adverse events were consistent across the Chinese and global populations. Two patients (10.0%) in this study achieved a partial response. CONCLUSION: In this study, tiragolumab plus atezolizumab was tolerable and demonstrated preliminary anti-tumor activity. There were no meaningful differences in the PK or safety of tiragolumab plus atezolizumab between the Chinese and global populations. CLINICAL TRIAL REGISTRATION NUMBER: China Clinical Trial Registry Identifier CTR20210219/YP42514. Date of registration 16 March 2021.

TP53 mitigates cisplatin resistance in non-small cell lung cancer by mediating the effects of resistant cell-derived exosome mir-424-5p.

Background: Cisplatin (DDP) is the principal agent used for chemotherapy in patients with non-small cell lung cancer (NSCLC). Nevertheless, DDP resistance is an essential cause for a worse prognosis of patient. Therefore, this study proposes to discover features of miR-424-5p in DDP resistance of NSCLC. Method: After exogenous modulation of miR-424-5p expression, A549 cell activity was measured using CCK-8 and flow cytometry. A549/DDP and A549/DDP-associated subcutaneous tumor model were constructed to investigate the effect of miR-424-5p on DDP resistance in NSCLC in vivo. TargetScan and JASPAR databases predicted the potential molecular mechanism of miR-424-5p. A549-and A549/DDP-derived exosomes were isolated and characterized using a transmission electron microscope and nanoparticle tracking analysis. Result: Overexpression of miR-424-5p facilitated proliferation and DDP resistance in A549 cells, and knockdown of miR-424-5p did the opposite. Knockdown of miR-424-5p enhanced DDP restriction on tumor weight and volume. Moreover, SOCS5 and SOCS56 (SOCS5/6) were downstream targets of miR-424-5p. miR-424-5p down-regulated SOCS5/6 expression to activate JAK2/STAT3 and PI3K/AKT pathways. Notably, tumor protein p53 (TP53) is a transcription factor for the miR-424-5p host gene, as confirmed by the dual-luciferase reporter gene. Cellular and animal experiments indicated that TP53 limited the regulatory function of miR-424-5p on NSCLC growth, DDP resistance, and related molecules. Interestingly, miR-424-5p was markedly enriched in A549/DDP cell-derived exosomes than in A549 cell-derived exosomes, and TP53 down-regulated miR-424-5p expression in A549/DDP cell-derived exosomes. Conclusion: DDP-resistant cell-derived exosome miR-424-5p contributes to NSCLC growth and DDP resistance by targeting SOCS5 and SOCS6 to activate JAK2/STAT3 and PI3K/AKT pathways, which are blocked by TP53.

Preparation and Application of Apatite-TiO2 Composite Opacifier: Preventing Titanium Glaze Yellowing through Pre-Combination.

In order to enhance the degree of binding reaction of TiO2 in titanium-containing ceramic glazes and prevent the reaction of its transformation into rutile to eliminate the yellowing phenomenon of the glaze surface, an apatite-TiO2 composite opacifier (ATO) was prepared through the mechanical grinding of hydroxyapatite and anatase TiO2. The properties, opacification mechanism, and yellowing inhibition of the prepared ceramic glazes were studied. The results show that the ATO is characterized by a uniform coating of TiO2 on the surface of the apatite and the formation of close chemical bonding between the apatite and TiO2. The ceramic glaze surface when using an ATO has a white appearance and excellent opacification performance. When an ATO was used, the L*, a*, and b* values of the glaze were 89.99, -0.85, and 3.37, respectively, which were comparable to those of a ZrSiO4 glaze (L*, a*, and b* were 88.24, -0.02, and 2.29, respectively). The opacification of the glaze was slightly lower than that of the TiO2 glaze (L* value was 92.13), but the appearance changed from yellow to the white of the TiO2 glaze (b* value was 9.18). The ceramic glaze layer when using an ATO mainly consists of titanite, glass phase, and a small amount of quartz, and the opacification mechanism is the crystallization of the generated titanite. ATOs can play an active role in solving the critical problem that arises when TiO2 replaces ZrSiO4 as an opacifier.

Neuroprotective effects of dipeptidyl peptidase 4 inhibitor on Alzheimer's disease: a narrative review.

Insulin resistance in brain and amyloidogenesis are principal pathological features of diabetes-related cognitive decline and development of Alzheimer's disease (AD). A growing body of evidence suggests that maintaining glucose under control in diabetic patients is beneficial for preventing AD development. Dipeptidyl peptidase 4 inhibitors (DDP4is) are a class of novel glucose-lowering medications through increasing insulin excretion and decreasing glucagon levels that have shown neuroprotective potential in recent studies. This review consolidates extant evidence from earlier and new studies investigating the association between DPP4i use, AD, and other cognitive outcomes. Beyond DPP4i's benefits in alleviating insulin resistance and glucose-lowering, underlying mechanisms for the potential neuroprotection with DPP4i medications were categorized into the following sections: (Ferrari et al., Physiol Rev, 2021, 101, 1,047-1,081): the benefits of DPP4is on directly ameliorating the burden of ß-amyloid plaques and reducing the formation of neurofibrillary tangles; DPP4i increasing the bioactivity of neuroprotective DPP4 substrates including glucagon-like peptide-1 (GLP-1), glucose-dependent insulinotropic peptide (GIP), and stromal-derived factor-1α (SDF-1α) etc.; pleiotropic effects of DPP4is on neuronal cells and intracerebral structure including anti-inflammation, anti-oxidation, and anti-apoptosis. We further revisited recently published epidemiological studies that provided supportive data to compliment preclinical evidence. Given that there remains a lack of completed randomized trials that aim at assessing the effect of DPP4is in preventing AD development and progression, this review is expected to provide a useful insight into DPP4 inhibition as a potential therapeutic target for AD prevention and treatment. The evidence is helpful for informing the rationales of future clinical research and guiding evidence-based clinical practice.

Value of Magnetic Resonance T1 Mapping in Evaluating the Early Response to Treatment for Rheumatoid Arthritis.

BACKGROUND: Rheumatoid Arthritis Magnetic Resonance Imaging Score (RAMRIS) is usually used for the semi-quantitative evaluation of joint changes in Rheumatoid Arthritis (RA). However, this method cannot evaluate early changes in bone marrow edema (BME). OBJECTIVE: To determine whether T1 mapping of wrist BME predicts early treatment response in RA. METHODS: This study prospectively enrolled 48 RA patients administered oral anti-rheumatic drugs. MRI of the most severely affected wrist was performed before and after 4 (48 patients) and 8 weeks of treatment (38 patients). Mean T1 values of BME in the lunate, triangular, and capitate bones; RAMRIS for each wrist; Erythrocyte-Sedimentation Rate (ESR); and 28-joint Disease Activity Score (DAS28)-ESR score were analyzed. Patients were divided into responders (4 weeks, 30 patients; 8 weeks, 32 patients) and non-responders (4 weeks, 18 patients; 8 weeks, 6 patients), according to EULAR response criteria. Receiver operating characteristic (ROC) curves were used to evaluate the efficacy of T1 values. RESULTS: ESR and DAS28-ESR were not correlated with T1 value and RAMRIS at each examination (P > 0.05). Changes in T1 value and DAS28-ESR relative to the baseline were moderately positively correlated with each other at 4 and 8 weeks (r = 0.555 and 0.527, respectively; P < 0.05). At 4 weeks, the change and rate of change in T1 value significantly differed between responders and non-responders (-85.63 vs. -19.92 ms; -12.89% vs. -2.81%; P < 0.05). The optimal threshold of the rate of change in T1 value at 4 weeks for predicting treatment response was -5.32% (area under the ROC curve, 0.833; sensitivity, 0.900; specificity, 0.667). CONCLUSION: T1 mapping provides a new imaging method for monitoring RA lesions; changes in wrist BME T1 values reflect early treatment response.

PB-LKS: a python package for predicting phage-bacteria interaction through local K-mer strategy.

Bacteriophages can help the treatment of bacterial infections yet require in-silico models to deal with the great genetic diversity between phages and bacteria. Despite the tolerable prediction performance, the application scope of current approaches is limited to the prediction at the species level, which cannot accurately predict the relationship of phages across strain mutants. This has hindered the development of phage therapeutics based on the prediction of phage-bacteria relationships. In this paper, we present, PB-LKS, to predict the phage-bacteria interaction based on local K-mer strategy with higher performance and wider applicability. The utility of PB-LKS is rigorously validated through (i) large-scale historical screening, (ii) case study at the class level and (iii) in vitro simulation of bacterial antiphage resistance at the strain mutant level. The PB-LKS approach could outperform the current state-of-the-art methods and illustrate potential clinical utility in pre-optimized phage therapy design.

Single-cell and Bulk RNA-Seq reveal angiogenic heterogeneity and microenvironmental features to evaluate prognosis and therapeutic response in lung adenocarcinoma.

Background: Angiogenesis stands as a pivotal hallmark in lung adenocarcinoma (LUAD), intricately shaping the tumor microenvironment (TME) and influencing LUAD progression. It emerges as a promising therapeutic target for LUAD, affecting patients' prognosis. However, its role in TME, LUAD prognosis, and its clinical applicability remain shrouded in mystery. Methods: We employed integrated single-cell and bulk transcriptome sequencing to unravel the heterogeneity of angiogenesis within LUAD cells. Through "consensus clustering", we delineated distinct angiogenic clusters and deciphered their TME features. "Monocle2" was used to unravel divergent trajectories within malignant cell subpopulations of LUAD. Additionally, regulon submodules and specific cellular communication patterns of cells in different angiogenic states were analyzed by "pyscenic" and "Cellchat" algorithms. The "univariate Cox" and "LASSO" algorithms were applied to build angiogenic prognostic models. Immunohistochemistry (IHC) on clinical samples validated the role of model factors in LUAD angiogenesis. We utilized CTRP 2.0 and PRISM databases for pinpointing sensitive drugs against lung adenocarcinoma. Results: Two clusters for the activation of angiogenesis were identified, with Cluster 1 showing a poor prognosis and a pro-cancerous TME. Three differentiated states of malignant epithelial LUAD cells were identified, which had different degrees of angiogenic activation, were regulated by three different regulon submodules, and had completely different crosstalk from other cells in TME. The experiments validate that SLC2A1 promotes angiogenesis in LUAD. ARS (Angiogenesis related score) had a high prognostic value; low ARSs showed immunotherapy benefits, whereas high ARSs were sensitive to 15 chemotherapeutic agents. Conclusion: The assessment of angiogenic clusters helps to determine the prognostic and TME characteristics of LUAD. Angiogenic prognostic models can be used to assess the prognosis, immunotherapeutic response, and chemotherapeutic drug sensitivity of LUAD.