Predict Patient Outcome Research Articles

The Use of AI in Enhancing Patient Monitoring Systems

Artificial Intelligence (AI) is transforming the healthcare landscape by automating complex tasks and enhancing patient safety. This paper examines AI’s role in detecting adverse events, improving clinical decision-making, and predicting patient outcomes. AI-based systems, particularly in medical imaging and diagnostics, are helping reduce errors and prevent unnecessary harm. However, challenges such as data quality, system transparency, and ethical concerns present hurdles to its widespread adoption. This review examines current literature on AI’s impact on patient safety, identifying the potential benefits and limitations. It also discusses future directions for AI-driven healthcare, emphasizing the need for robust regulatory frameworks and ethical considerations. Keywords: AI in healthcare, patient safety, clinical decision support, medical imaging, predictive algorithms.

The Use of AI in Predicting Patient Outcomes

Artificial intelligence (AI) has emerged as a revolutionary tool in healthcare, significantly improving the ability to predict patient outcomes. From chronic disease management to personalized treatments, AI can analyze vast datasets and provide timely, accurate predictions that guide clinical decision-making. Traditional statistical models have been useful but often limited in handling complex data relationships. AI, through machine learning (ML) and deep learning (DL), transcends these limitations, offering improved predictions across diseases like cancer, tuberculosis, and HIV/AIDS. This paper investigates current AI methods, advantages, challenges, and ethical considerations associated with its use in healthcare, highlighting the transformative potential of AI in outcome prediction while addressing concerns around data privacy, regulatory constraints, and model transparency. Keywords: Artificial Intelligence, Machine Learning, Patient Outcomes, Clinical Decision Support Systems, Predictive Analytics.

Improving performance in colorectal cancer histology decomposition using deep and ensemble machine learning

In routine colorectal cancer management, histologic samples stained with hematoxylin and eosin are commonly used. Nonetheless, their potential for defining objective biomarkers for patient stratification and treatment selection is still being explored. The current gold standard relies on expensive and time-consuming genetic tests. However, recent research highlights the potential of convolutional neural networks (CNNs) to facilitate the extraction of clinically relevant biomarkers from these readily available images. These CNN-based biomarkers can predict patient outcomes comparably to golden standards, with the added advantages of speed, automation, and minimal cost. The predictive potential of CNN-based biomarkers fundamentally relies on the ability of CNNs to accurately classify diverse tissue types from whole slide microscope images. Consequently, enhancing the accuracy of tissue class decomposition is critical to amplifying the prognostic potential of imaging-based biomarkers. This study introduces a hybrid deep transfer learning and ensemble machine learning model that improves upon previous approaches, including a transformer and neural architecture search baseline for this task. We employed a pairing of the EfficientNetV2 architecture with a random forest classification head. Our model achieved 96.74% accuracy (95% CI: 96.3%-97.1%) on the external test set and 99.89% on the internal test set. Recognizing the potential of these models in the task, we have made them publicly available.

Utility of combining frailty and comorbid disease indices in predicting outcomes following craniotomy for adult primary brain tumors: A mixed-effects model analysis using the nationwide readmissions database

ObjectiveThe escalating healthcare expenditures in the United States, particularly in neurosurgery, necessitate effective tools for predicting patient outcomes and optimizing resource allocation. This study explores the utility of combining frailty and comorbidity indices, specifically the Johns Hopkins Adjusted Clinical Groups (JHACG) frailty index and the Elixhauser Comorbidity Index (ECI), in predicting hospital length of stay (LOS), non-routine discharge, and one-year readmission in patients undergoing craniotomy for benign and malignant primary brain tumors. MethodsLeveraging the Nationwide Readmissions Database (NRD) for 2016–2019, we analyzed data from 645 patients with benign and 30,991 with malignant tumors. Frailty, ECI, and frailty + ECI were assessed as predictors using generalized linear mixed-effects models. Receiver operating characteristic (ROC) curves evaluated predictive performance. ResultsPatients in the benign tumor cohort had a mean LOS of 8.1 ± 15.1 days, and frailty + ECI outperformed frailty alone in predicting non-routine discharge (AUC 0.829 vs. 0.820, p = 0.035). The malignant tumor cohort patients had a mean LOS of 7.9 ± 9.1 days. In this cohort, frailty + ECI (AUC 0.821) outperformed both frailty (AUC 0.744, p < 0.0001) and ECI alone (AUC 0.809, p < 0.0001) in predicting hospital LOS. Frailty + ECI (AUC 0.831) also proved superior to frailty (AUC 0.809, p < 0.0001) and ECI alone (AUC 0.827, p < 0.0001) in predicting non-routine discharge location for patients with malignant tumors. All indices performed comparably to one another as a predictor of readmission in both cohorts. ConclusionThis study highlights the synergistic predictive capacity of frailty + ECI, especially in malignant tumor cases, and further suggests that comorbid diseases may greatly influence perioperative outcomes more than frailty. Enhanced risk assessment could aid clinical decision-making, patient counseling, and resource allocation, ultimately optimizing patient outcomes.

A deep learning approach for cervical cord injury severity determination through axial and sagittal magnetic resonance imaging segmentation and classification.

Cross-sectional Database Study. While the American Spinal Injury Association (ASIA) Impairment Scale is the standard for assessing spinal cord injuries (SCI), it has limitations due to subjectivity and impracticality. Advances in machine learning (ML) and image recognition have spurred research into their use for outcome prediction. This study aims to analyze deep learning techniques for identifying and classifying cervical SCI severity from MRI scans. The study included patients with traumatic and nontraumatic cervical SCI admitted from 2019 to 2022. MRI images were labeled by two senior resident physicians. A deep convolutional neural network was trained using axial and sagittal cervical MRI images from the dataset. Model performance was assessed using Dice Score and IoU to measure segmentation accuracy by comparing predicted and ground truth masks. Classification accuracy was evaluated with the F1 Score, balancing false positives and negatives. In the axial spinal cord segmentation, we achieved a Dice score of 0.94 for and IoU score of 0.89. In the sagittal spinal cord segmentation, we obtained Dice score up to 0.9201 and IoU scores up to 0.8541. The model for axial image score classification gave a satisfactory result with an F1 score of 0.72 and AUC of 0.79. Our models successfully identified cervical SCI on T2-weighted MR images with satisfactory performance. Further research is needed to develop more advanced models for predicting patient outcomes in SCI cases.

The Role of Artificial Intelligence in Healthcare: Applications, Challenges, and Ethical Considerations

Artificial intelligence (AI) has emerged as a transformative technology with the potential to revolutionize healthcare delivery, from diagnosis and treatment to patient care and administrative tasks. the applications, challenges, and ethical considerations surrounding the role of AI in healthcare. It discusses how AI algorithms can analyze vast amounts of medical data to assist healthcare professionals in making accurate diagnoses, predicting patient outcomes, and personalizing treatment plans. Additionally, it examines the challenges associated with implementing AI in healthcare, such as data privacy concerns, algorithm bias, and regulatory hurdles. Furthermore, it addresses ethical considerations, including transparency, accountability, and the impact of AI on patient-provider relationships. Despite these challenges, AI holds tremendous promise for improving healthcare efficiency, accessibility, and quality, provided that stakeholders address these concerns and harness AI's potential responsibly.

Optimal therapeutic strategies for hepatic metachronous oligometastatic nasopharyngeal carcinoma: Insights from a retrospective study.

Hepatic metachronous oligometastatic nasopharyngeal carcinoma (hmoNPC) exhibits distinct clinical characteristics compared to other types of metastatic NPC. We investigated the optimal therapy for hmoNPC. 160 patients with hmoNPC treated in Sun Yat-sen University Cancer Center between 2010 and 2021 were retrospectively recruited. A total of 56 patients were classified into the local therapy (LT) cohort, 23 into the systemic therapy (ST) cohort and 81 into the combination therapy (LT + ST) cohort. The median PFS was 7.9 months (95% confidence interval [CI]: 4.1-11.9 months) in the LT cohort, 15.5 months (95% CI: 10.5-32.3 months) in the ST cohort, and 31.3 months (95% CI: 20.3 to NA months) in the LT + ST cohort. The median OS was 41.1 months (95% CI: 30.0-54.0 months) in the LT cohort, 50.4 months (95% CI: 41.5 to NA months) in the ST cohort and not reached (NR) (95% CI: 77.3 to NA months) in the LT + ST cohort. Cox analysis was used to construct nomograms to predict patient outcomes. Among patients with no evidence of disease status after LT, the prognosis was significantly better in the LT + ST cohort than LT cohort (median PFS: NR [95% CI: 29.0 to NA months] vs. 20.0 months [95% CI: 10.4 to NA months]). More survival benefits were achieved with platinum-based chemotherapy than oral monotherapy (median PFS: NR [95% CI: 21.7 to NA months] vs. 17.2 months [95% CI: 10.2 to NA months]). Fewer postoperative early progression events were observed in neoadjuvant chemotherapy cohort than in adjuvant chemotherapy cohort (2.78% vs. 18.81%, P = .013). In conclusion, combining neoadjuvant platinum-based chemotherapy and local therapy was the best strategy for patients with hmoNPC.

Comparative analysis of four upper gastrointestinal bleeding scoring systems for predicting multiple outcomes: an observational study in the emergency department

Objective: Numerous scoring systems have been developed to assess the risk associated with upper gastrointestinal bleeding (UGIB), and several studies have investigated their comparative accuracy in predicting patient outcomes. This study was undertaken to compare four well-known scoring systems, namely the pre-endoscopy Rockall score, full Rockall score, Glasgow-Blatchford Bleeding score (GBS), and AIMS65, with the aim of predicting five distinct outcomes in cases of non-variceal UGIB. Methods: This prospective observational study was conducted focusing on adult patients with UGIB presenting to the emergency department (ED). The primary endpoints of this study included in-hospital mortality, the need for re-endoscopy, the requirements for packed red blood cell (PRBC) transfusion, massive transfusion, and one-month rebleeding. Results: A total number of 320 patients were enrolled, with 44 (13·75%) in-hospital deaths. Based on the area under the curves (AUC), while certain scores outperformed others in specific outcome prediction, the AIMS65 scoring system demonstrated superior predictive capability for both in-hospital mortality (0.91) and massive transfusion (0.71). Regarding PRBC transfusion requirements, both AIMS65 and GBS exhibited similar predictive capacities (AUC=0.67 and 0.68, respectively). In terms of re-endoscopy and one-month rebleeding, the GBS scoring system displayed slightly better performance compared to the other systems (AUC=0.61 and 0.63, respectively). In the composite outcome, all scores had significant associations, and among them, the AIMS-65 score had the highest AUC (0.76). Conclusion: The AIMS65 scoring system was the most reliable tool for predicting in-hospital mortality and, to a lesser extent, massive transfusion requirements, while GBS and AIMS65 could be moderately and cautiously relied on for preparations regarding the need for PRBC transfusion.

Spatial heterogeneity and prognostic significance of TAMs and TILs infiltrates in different staging esophageal squamous carcinoma

BackgroundThe prognostic value and clinical relevance of tumor-infiltrating lymphocytes (TILs) and tumor-associated macrophages (TAMs) in esophageal squamous cell carcinoma (ESCC) remain unclear. AimsTo investigate the prognostic value and functional involvement of TILs in ESCC. MethodsWe included 40 patients across different stages of ESCC from Xinjiang. Multiplex fluorescent immunohistochemistry characterized TILs and TAMs. TILs in different tumor regions were quantified and correlated with overall survival (OS) using log-rank test and Cox regression analyses. ResultsInvasive ESCC exhibited increased CD4 T cells and Tregs compared to carcinoma in situ, with a higher Tregs/CD4 T cells ratio (p < 0.05). TAMs, primarily in stromal regions, were significantly associated with Foxp3+ cells (p < 0.05). Higher infiltration of stromal TAMs and a higher CD4/CD8 T cells ratio correlated with poorer OS, while a higher CD8 T/Foxp3+ cells ratio indicated better survival. Multivariate Cox analysis revealed TNM stage, tumor length, and stromal CD4/CD8 T cells ratio as independent prognostic factors (p < 0.05). An immune prognostic risk score-based nomogram was constructed to predict patient outcomes. ConclusionsThe spatial distribution and abundance of TILs significantly correlated with prognosis, providing a useful immune classification for ESCC.

G0 arrest gene patterns to predict the prognosis and drug sensitivity of patients with lung adenocarcinoma.

G0 arrest (G0A) is widely recognized as a crucial factor contributing to tumor relapse. The role of genes related to G0A in lung adenocarcinoma (LUAD) was unclear. This study aimed to develop a gene signature based on for LUAD patients and investigate its relationship with prognosis, tumor immune microenvironment, and therapeutic response in LUAD. We use the TCGA-LUAD database as the discovery cohort, focusing specifically on genes associated with the G0A pathway. We used various statistical methods, including Cox and lasso regression, to develop the model. We validated the model using bulk transcriptome and single-cell transcriptome datasets (GSE50081, GSE72094, GSE127465, GSE131907 and EMTAB6149). We used GSEA enrichment and the CIBERSORT algorithm to gain insight into the annotation of the signaling pathway and the characterization of the tumor microenvironment. We evaluated the response to immunotherapy, chemotherapy, and targeted therapy in these patients. The expression of six genes was validated in cell lines by quantitative real-time PCR (qRT-PCR). Our study successfully established a six-gene signature (CHCHD4, DUT, LARP1, PTTG1IP, RBM14, and WBP11) that demonstrated significant predictive power for overall survival in patients with LUAD. It demonstrated independent prognostic value in LUAD. To enhance clinical applicability, we developed a nomogram based on this gene signature, which showed high reliability in predicting patient outcomes. Furthermore, we observed a significant association between G0A-related risk and tumor microenvironment as well as drug susceptibility, highlighting the potential of the gene signature to guide personalized treatment strategies. The expression of six genes were significantly upregulated in the LUAD cell lines. This signature holds the potential to contribute to improved prognostic prediction and new personalized therapies specifically for LUAD patients.

Comprehensive analysis identifies ubiquitin ligase FBXO42 as a tumor-promoting factor in neuroblastoma

Neuroblastoma, the deadliest solid tumor in children, exhibits alarming mortality rates, particularly among high-risk cases. To enhance survival rates, a more precise risk stratification for patients is imperative. Utilizing proteomic data from 34 cases with or without N-Myc amplification, we identified 28 differentially expressed ubiquitination-related proteins (URGs). From these, a prognostic signature comprising 6 URGs was constructed. A nomogram incorporating clinical-pathological parameters yielded impressive AUC values of 0.88, 0.93, and 0.95 at 1, 3, and 5 years, respectively. Functional experiments targeting the E3 ubiquitin ligase FBXO42, a component of the prognostic signature, revealed its TP53-dependent promotion of neuroblastoma cell proliferation. In conclusion, our ubiquitination-related prognostic model robustly predicts patient outcomes, guiding clinical decisions. Additionally, the newfound pro-proliferative role of FBXO42 offers a novel foundation for understanding the molecular mechanisms of neuroblastoma.

The role of serum uric acid in predicting left ventricular function and outcomes in patients with STEMI following PCI

Objectives: Percutaneous coronary intervention (PCI) is the preferred treatment for ST-elevation myocardial infarction (STEMI); however, reduced left ventricular ejection fraction (LVEF) remains a critical issue post-PCI. Serum uric acid (SUA) has emerged as a biomarker linked to cardiovascular events and may be associated with LVEF. This study aims to investigate the relationship between SUA levels and LVEF in STEMI patients treated with primary PCI and identify potential biomarkers for predicting patient outcomes. Methods: This prospective study was conducted at Afshar Hospital, Shahid Sadoughi University of Medical Sciences of Yazd, from September 2019 to March 2021. Participants included 254 consecutive ST-elevation myocardial infarction (STEMI) patients undergoing primary PCI. Exclusion criteria were contraindications for angiography, history of myocardial infarction, PCI, CABG, NSTEMI, thrombolytic treatment, kidney diseases, gout, or alcoholism. Demographic data, clinical history, SUA levels, and echocardiographic parameters were collected. Statistical analysis was performed using SPSS version 20.0, with significance at p&lt;0.05. Results: 81.49% of patients were male, with a mean age of 57.7±11.47 years. Hypertension was the most prevalent condition (42.9%). SUA levels did not significantly differ between patients with LVEF below and above 40% (p=0.39). However, smoking was significantly associated with reduced LVEF (p&lt;0.001). A weak negative correlation between SUA and LVEF was observed in males (p=0.012) but not in females (p=0.097). Overall, a weak negative correlation between SUA and LVEF was statistically significant (p=0.05). Conclusions: Our findings revealed a weak but statistically significant overall negative correlation between SUA and EF. Further research with larger, more diverse populations is necessary to elucidate the relationship between SUA and cardiac function in STEMI patients.

Bioinformatics analysis of the tumor microenvironment in melanoma - Constructing a prognostic model based on CD8+ T cell-related genes: An observational study.

This research endeavor seeks to explore the microenvironment of melanoma tumors and construct a prognostic model by focusing on genes specific to CD8+ T cells. The single-cell sequencing data of melanoma underwent processing with the Seurat package, subsequent to which cell communication network analysis was conducted using the iTALK package and transcription factor analysis was performed using the SCENIC package. Univariate COX and LASSO regression analyses were utilized to pinpoint genes linked to the prognosis of melanoma patients, culminating in the creation of a prognostic model through multivariate COX analysis. The model was validated using the GSE65904 and GSE35640 datasets. Multi-omics analysis was conducted utilizing the maftools, limma, edgeR, ChAMP, and clusterProfiler packages. The examination of single-cell sequencing data revealed the presence of 8 cell types, with the transcription factors RFXAP, CLOCK, MGA, RBBP, and ZNF836 exhibiting notably high expression levels in CD8+ T cells as determined by the SCENIC package. Utilizing these transcription factors and their associated target genes, a prognostic model was developed through COX and LASSO analyses, incorporating the genes GPR171, FAM174A, and BPI. This study validated the model with independent datasets and conducted additional analysis involving multi-omics and immune infiltration to identify a more favorable prognosis for patients in the low-risk group. The findings provide valuable insights into the tumor microenvironment of melanoma and establish a reliable prognostic model. The integration of multi-omics and immune infiltration analyses enhances our understanding of the pathogenesis of melanoma. The identification of specific genes holds promise as potential biomarkers for individuals with melanoma, serving as important indicators for predicting patient outcomes and determining their response to immunotherapy.

MiRNA Dysregulation in Brain Injury: An In Silico Study to Clarify the Role of a MiRNA Set.

The identification of specific circulating miRNAs has been proposed as a valuable tool for elucidating the pathophysiology of brain damage or injury and predicting patient outcomes. This study aims to apply several bioinformatic tools in order to clarify miRNA interactions with potential genes involved in brain injury, emphasizing the need of using a computational approach to determine the most likely correlations between miRNAs and target genes. Specifically, this study centers on elucidating the roles of miR-34b, miR-34c, miR-135a, miR-200c, and miR-451a. After a careful evaluation of different software available (analyzing the strengths and limitations), we applied three tools, one to perform an analysis of the validated targets (miRTarBase), and two to evaluate functional annotations (miRBase and TAM 2.0). Research findings indicate elevated levels of miR-135a and miR-34b in patients with traumatic brain injury (TBI) within the first day post-injury, while miR-200c and miR-34c were found to be upregulated after 7 days. Moreover, miR-451a and miR-135a were found overexpressed in the serum, while miRNAs 34b, 34c, and 200c, had lower serum levels at baseline post brain injury. This study emphasizes the use of computational methods in determining the most likely relationships between miRNAs and target genes by investigating several bioinformatic techniques to elucidate miRNA interactions with potential genes. Specifically, this study focuses on the functions of miR-34b, miR-34c, miR-135a, miR-200c, and miR-451a, providing an up-to-date overview and suggesting future research directions for identifying theranomiRNAs related to brain injury, both at the tissue and serum levels.

Impact of perioperative hemoglobin-related parameters on clinical outcomes in patients with spinal metastases: identifying key markers for blood management.

Patients with spinal metastases undergoing surgical treatment face challenges related to preoperative anemia, intraoperative blood loss, and frailty, emphasizing the significance of perioperative blood management. This retrospective analysis aimed to assess the correlation between hemoglobin-related parameters and outcomes, identifying key markers to aid in blood management. A retrospective review was performed to identify patients who underwent surgical treatment for spinal metastases. Hb-related parameters, including baseline Hb, postoperative nadir Hb, predischarge Hb, postoperative nadir Hb drift, and predischarge Hb drift (both in absolute values and percentages) were subjected to univariate and multivariate analyses. These analyses were conducted in conjunction with other established variables to identify independent markers predicting patient outcomes. The outcomes of interest were postoperative short-term (6-week) mortality, long-term (1-year) mortality, and postoperative 30-day morbidity. A total of 289 patients were included. Our study demonstrated that predischarge Hb (OR 0.62, 95% CI 0.44-0.88, P = 0.007) was an independent prognostic factor of short-term mortality, while baseline Hb (OR 0.76, 95% CI 0.66-0.88, P < 0.001) was identified as an independent prognostic factor of long-term mortality. Additionally, nadir Hb drift (OR 0.82, 95% CI 0.70-0.97, P = 0.023) was found to be an independent prognostic factor for postoperative 30-day morbidity. This study demonstrated that predischarge Hb, baseline Hb, and nadir Hb drift are prognostic factors for outcomes. These findings provide a foundation for precise blood management strategies. It is crucial to consider Hb-related parameters appropriately, and prospective intervention studies addressing these markers should be conducted in the future.

Glycolysis-related genes predict prognosis and indicate immune microenvironment features in gastric cancer.

Gastric cancer (GC) is a major contributor to cancer-related mortality. Glycolysis plays a pivotal role in tumor microenvironment (TME) reprogramming. In this research, the functions of glycolysis-associated genes (GRGs) were evaluated to predict the outcome and reveal the characteristics of the immune microenvironment in individuals with stomach cancer. The Cancer Genome Atlas (TCGA)-stomach adenocarcinoma (STAD) cohort provided gene expression and clinical data for gastric cancer (GC) patients, which were further authenticated using datasets sourced from the Gene Expression Omnibus (GEO). By referencing the Molecular Signatures Database (MSigDB), a total of 326 GRGs were pinpointed. The various subtypes of GC were outlined through consensus clustering, derived from the expression patterns of these GRGs. Utilizing multivariate Cox regression analysis, a multigene risk score model was formulated. Both the CIBERSORT and ESTIMATE algorithms played a pivotal role in assessing the immune microenvironment. To delve into the biological functions of the key genes, wound healing, transwell invasion, and MTT assays were conducted. Based on the expression patterns of GRGs, patients were categorized into two distinct groups: the metabolic subtype, designated as cluster A, and the immune subtype, labeled as cluster B. Patients belonging to cluster B exhibited a poorer prognosis. A prognostic risk score model, formulated upon the expression levels of six key GRGs - ME1, PLOD2, NUP50, CXCR4, SLC35A3, and SRD35A3 - emerged as a viable tool for predicting patient outcomes. The downregulation of CXCR4 notably diminished the glycolytic capacity of gastric cancer (GC) cells, alongside their migratory, invasive, and proliferative capabilities. Intriguingly, despite the adverse prognostic implications associated with both the immune subtype (cluster B) and the high-risk cohort, these groups exhibited a favorable immune microenvironment coupled with elevated expression of immune checkpoint genes. Our investigations revealed a positive correlation between high CXCR4 expression and low ME1 expression with the infiltration of CD8+ T cells, as well as an enhanced responsiveness to treatment with an anti-PD-1 immune checkpoint inhibitor. In this study, we discovered that the expression profiles of GRGs hold the potential to forecast the prognosis of gastric cancer (GC) patients, thereby possibly aiding in clinical treatment decision-making.

Comparison of Four Severity Assessment Scoring Systems in Critically Ill Patients for Predicting Patient Outcomes: A Prospective Observational Study From a Single Tertiary Center in Central India.

Background and aim A variety of scoring systems are employed in intensive care units (ICUs) with the objective of predicting patient morbidity and mortality. The present study aimed to compare four different severity assessment scoring systems, namely, Acute Physiology and Chronic Health Evaluation II (APACHE II), Rapid Emergency Medicine Score (REMS), Sequential Organ Failure Assessment (SOFA), and Simplified Acute Physiologic Score II (SAPS II) to predict prognosis of all patients admitted to a mixed medical ICU of a tertiary care teaching hospital in central India. Methods The prospective observational study included 1136 patients aged 18 years or more, admitted to the mixed medical ICU. All patients underwent severity assessment using the four scoring systems, namely APACHE II, SOFA, REMS, and SAPS II, after admission. Predicted mortality was calculated from each of the scores and actual patient outcomes were noted. Receiver operating curve analysis was undertaken to identify the cut-off value of individual scoring systems for predicting mortality with optimum sensitivity and specificity. Calibration and discrimination were employed to ascertain the validity of each scoring model. Bivariate and multivariable logistic regression analyses among the study participants were conducted to identify the best scoring system, after adjusting for potential confounders. Results Final analysis was done on 957 study participants (mean (±SD) age-58.4 (±12.9) years; males-62.2%). The mortality rate was 14.7%. APACHE II, SOFA, SAPS II, and REMS scores were significantly higher among the non-survivors as compared to the survivors (p<0.05). SAPS II was found to have the highest AUC of 0.981 (p<0.001). SAPS II score >58 had 93.6% sensitivity, 94.1% specificity, 73.3% PPV, 98.8% NPV, and 94.0% diagnostic accuracy in predicting mortality. This scoring system also had the best calibration. Binary logistic regression showed that all four scoring systems were significantly associated with ICU mortality. After adjusting for each other, only SAPS II remained significantly associated with ICU mortality. Conclusion Both SAPS II and APACHE II were observed to have good calibration and discriminatory power; however, SAPS II had the best prediction power suggesting that it may be a useful tool for clinicians and researchers in assessing the severity of illness and mortality risk in critically ill patients.

MiRNA expression as outcome predictor in pediatric AML: systematic evaluation of a new model

Accurately predicting patient outcomes is essential for optimizing treatment and improving outcomes in pediatric acute myeloid leukemia (AML). In recent years, microRNAs have emerged as a promising prognostic marker, with a growing body of evidence supporting their potential predictive value. We systematically reviewed all previous studies that have analyzed the expression of microRNAs as predictors of survival in pediatric AML and found 16 microRNAs and 4 microRNA signatures previously proposed as predictors of survival. We then used a public access cohort of 1414 pediatric AML patients from the TARGET project to develop a new predictive model using penalized lasso Cox regression based on microRNA expression. Here we propose a new score based on a 37-microRNA signature that is associated with AML and is able to predict survival more accurately than previous microRNA-based methods.

Unlocking the potential of HHLA2: identifying functional immune infiltrating cells in the tumor microenvironment and predicting clinical outcomes in laryngeal squamous cell carcinoma

BackgroundHHLA2 (human endogenous retrovirus-H long terminal repeat-associating protein 2) represents a recently identified member of the B7 immune checkpoint family, characterized by limited expression in normal tissues but notable overexpression in various cancer types. Nevertheless, the precise function and interaction with immune cells remain poorly understood, particularly in laryngeal squamous cell carcinoma (LSCC). This investigation endeavored to elucidate the biological significance of HHLA2 within the tumor microenvironment of human LSCC tissues and delineate the clinical relevance and functional roles of HHLA2 in LSCC pathogenesis.MethodsThrough multiplexed immunohistochemistry analyses conducted on tissue microarrays sourced from LSCC patients (n = 72), the analysis was executed to assess the expression levels of HHLA2, density and spatial patterns of CD68+HLA-DR+CD163− (M1 macrophages), CTLA-4+CD4+FoxP3+ (CTLA-4+Treg cells), CTLA-4+CD4+FoxP3− (CTLA-4+Tcon cells), exhausted CD8+T cells, and terminally exhausted CD8+T cells in LSCC tissues. Survival analysis was conducted to evaluate the prognostic significance of HHLA2 and these immune checkpoints or immune cell populations, employing COX regression analysis to identify independent prognostic factors.ResultsKaplan–Meier (K–M) survival curves revealed a significant association between HHLA2 expression and overall survival (OS) in LSCC. Elevated levels of HHLA2 were linked to reduced patient survival, indicating its potential as a prognostic marker (HR: 3.230, 95%CI 0.9205–11.34, P = 0.0067). Notably, increased infiltration of CD68+ cells (total macrophages), STING+CD68+HLA-DR+CD163− (STING+M1 macrophages), CTLA-4+CD4+FoxP3+, CTLA-4+CD4+FoxP3−, PD-1+LAG-3+CD8+T cells, and PD-1+LAG-3+TIM-3+CD8+T cells strongly linked to poorer survival outcomes (P < 0.05). A discernible trend was observed between the levels of these immune cell populations, STING+CD68+ (STING+ total macrophages), CD68+HLA-DR+CD163−, STING+CD68+CD163+HLA-DR− (STING+M2 macrophages), PD-1+LAG-3−CD8+T cells, PD-1+TIM-3+CD8+T cells, and PD-1+LAG-3+TIM-3−CD8+T cells and prognosis. Importantly, multivariate COX analysis identified HHLA2 as an independent predictive factor for OS in LSCC patients (HR = 3.86, 95%CI 1.08–13.80, P = 0.038). This underscored the potential of HHLA2 as a critical marker for predicting patient outcomes in LSCC.ConclusionsHHLA2 emerged as a detrimental prognostic biomarker for assessing OS in LSCC patients. Relative to other immune checkpoints, HHLA2 exhibited heightened predictive efficacy for the prognosis of LSCC patients.

Comprehensive pancancer analysis reveals that LPCAT1 is a novel predictive biomarker for prognosis and immunotherapy response.

Lysophosphatidylcholine acyltransferase 1 (LPCAT1) is a crucial enzyme involved in phospholipid metabolism and is essential for maintaining the structure and functionality of biofilms. However, a comprehensive examination of the role of LPCAT1 across various cancer types is lacking. Multiple public databases have been utilized to examine LPCAT1 expression, genetic alterations, methylation, prognosis, biological function, and its relationship with antitumor immunity in different cancer types. The function of LPCAT1 in glioma, breast cancer and liver cancer cells was further verified using in vitro experiments. Our research indicated that LPCAT1 is upregulated in various cancers and is accompanied by a wide range of amplification mutations. Higher LPCAT1 expression was associated with poorer prognosis across multiple cancers. Further in vitro experiments demonstrated that interfering with LPCAT1 expression increased apoptosis in glioma, breast cancer and liver cancer cells and concurrently suppressed their proliferation and migration. Functional enrichment analysis revealed that LPCAT1-associated genes were primarily enriched in immune and cancer progression pathways, such as the JAK/STAT, MYC, and EMT, etc. Moreover, LPCAT1 expression was closely associated with immune cell infiltration and immune checkpoint-related gene expression. Interestingly, LPCAT1 expression levels were generally higher in patients in the immunotherapy response group. The combination of LPCAT1 and PDL1 serves as an effective predictor of immunotherapy response. In conclusion, LPCAT1 is involved in immune regulation and tumor progression and holds promise as a biomarker for predicting patient outcomes and immunotherapy efficacy.