Predict Patient Response Research Articles

Quantitative SPECT/CT Metrics in Early Prediction of [177Lu]Lu-DOTATATE Treatment Response in Gastroenteropancreatic Neuroendocrine Tumor Patients.

Our objective is to explore quantitative imaging markers for early prediction of treatment response in patients with gastroenteropancreatic neuroendocrine tumors (GEP-NETs) undergoing [177Lu]Lu-DOTATATE therapy. By doing so, we aim to enable timely switching to more effective therapies in order to prevent time-resource waste and minimize toxicities. Methods: Patients diagnosed with unresectable or metastatic, progressive, well-differentiated, receptor-positive GEP-NETs who received 4 sessions of [177Lu]Lu-DOTATATE were retrospectively selected. Using SPECT/CT images taken at the end of treatment sessions, we counted all visible tumors and measured their largest diameters to calculate the tumor burden score (TBS). Up to 4 target lesions were selected and semiautomatically segmented. Target lesion peak counts and spleen peak counts were measured, and normalized peak counts were calculated. Changes in TBS (ΔTBS) and changes in normalized peak count (ΔnPC) throughout treatment sessions in relation to the first treatment session were calculated. Treatment responses were evaluated using third-month CT and were binarized as progressive disease (PD) or non-PD. Results: Twenty-seven patients were included (7 PD, 20 non-PD). Significant differences were observed in ΔTBSsecond-first, ΔTBSthird-first, and ΔTBSfourth-first (where second-first, third-first, and fourth-first denote scan number between the second and first, third and first, and fourth and first [177Lu]Lu-DOTATATE treatment cycles), respectively) between the PD and non-PD groups (median, 0.043 vs. -0.049, 0.08 vs. -0.116, and 0.109 vs. -0.123 [P = 0.023, P = 0.002, and P < 0.001], respectively). ΔnPCsecond-first showed significant group differences (mean, -0.107 vs. -0.282; P = 0.033); ΔnPCthird-first and ΔnPCfourth-first did not reach statistical significance (mean, -0.122 vs. -0.312 and -0.183 vs. -0.405 [P = 0.117 and 0.067], respectively). At the optimal threshold, ΔTBSfourth-first exhibited an area under the curve (AUC) of 0.957, achieving 100% sensitivity and 80% specificity. ΔTBSsecond-first and ΔTBSthird-first reached AUCs of 0.793 and 0.893, sensitivities of 71.4%, and specificities of 85% and 95%, respectively. ΔnPCsecond-first, ΔnPCthird-first, and ΔnPCfourth-first showed AUCs of 0.764, 0.693, and 0.679; sensitivities of 71.4%, 71.4%, and 100%; and specificities of 75%, 70%, and 35%, respectively. Conclusion: ΔTBS and ΔnPC can predict [177Lu]Lu-DOTATATE response by the second treatment session.

Radiomics analysis of cerebral blood flow suggests a possible link between perfusion homogeneity and poor glioblastoma multiforme prognosis

Objectives. This study investigates the association between cerebral blood flow (CBF) and overall survival (OS) in glioblastoma multiforme (GBM) patients receiving chemoradiation. Identifying CBF biomarkers could help predict patient response to this treatment, facilitating the development of personalized therapeutic strategies. Materials and Methods. This retrospective study analyzed CBF data from dynamic susceptibility contrast (DSC) MRI in 30 newly diagnosed GBM patients (WHO grade IV). Radiomics features were extracted from CBF maps, tested for robustness, and correlated with OS. Kaplan-Meier analysis was used to assess the predictive value of radiomic features significantly associated with OS, aiming to stratify patients into groups with distinct post-treatment survival outcomes. Results. While mean relative CBF and CBV failed to serve as independent prognostic markers for OS, the prognostic potential of radiomic features extracted from CBF maps was explored. Ten out of forty-three radiomic features with highest intraclass correlation coefficients (ICC > 0.9), were selected for characterization. While Correlation and Zone Size Variance (ZSV) features showed significant OS correlations, indicating prognostic potential, Kaplan-Meier analysis did not significantly stratify patients based on these features. Visual analysis of the graphs revealed a predominant association between the identified radiomic features and OS under two years. Focusing on this subgroup, Correlation, ZSV, and Gray-Level Nonuniformity (GLN) emerged as significant, suggesting that a lack of heterogeneity in perfusion patterns may be indicative of a poorer outcome. Kaplan-Meier analysis effectively stratified this cohort based on the features mentioned above. Receiver operating characteristic (ROC) analysis further validated their prognostic value, with ZSV demonstrating the highest sensitivity and specificity (0.75 and 0.85, respectively). Conclusion. Our findings underscored radiomics features sensitive to CBF heterogeneity as pivotal predictors for patient stratification. Our results suggest that these markers may have the potential to identify patients who are unlikely to benefit from standard chemoradiation therapy.

Characterization of driver mutations identifies gene signatures predictive of prognosis and treatment sensitivity in multiple myeloma.

In multiple myeloma (MM), while frequent mutations in driver genes are crucial for disease progression, they traditionally offer limited insights into patient prognosis. This study aims to enhance prognostic understanding in MM by analyzing pathway dysregulations in key cancer driver genes, thereby identifying actionable gene signatures. We conducted a detailed quantification of mutations and pathway dysregulations in 10 frequently mutated cancer driver genes in MM to characterize their comprehensive mutational impacts on the whole transcriptome. This was followed by a systematic survival analysis to identify significant gene signatures with enhanced prognostic value. Our systematic analysis highlighted 2 significant signatures, TP53 and LRP1B, which notably outperformed mere mutation status in prognostic predictions. These gene signatures remained prognostically valuable even when accounting for clinical factors, including cytogenetic abnormalities, the International Staging System (ISS), and its revised version (R-ISS). The LRP1B signature effectively distinguished high-risk patients within low/intermediate-risk categories and correlated with significant changes in the tumor immune microenvironment. Additionally, the LRP1B signature showed a strong association with proteasome inhibitor pathways, notably predicting patient responses to bortezomib and the progression from monoclonal gammopathy of unknown significance to MM. Through a rigorous analysis, this study underscores the potential of specific gene signatures in revolutionizing the prognostic landscape of MM, providing novel clinical insights that could influence future translational oncology research.

380. GROWTH CHARACTERISTICS OF ESOPHAGEAL PATIENT DERIVED ORGANOIDS REFLECT CLINICOPATHOLOGIC VARIABLES

Abstract Background Esophageal patient derived organoids (PDOs) are three-dimensional cell culture models derived from esophageal cancer tissue. We have previously shown that PDOs recapitulate the genetic and morphologic characteristics of the primary tumour, and that PDOs may predict patient responses to neoadjuvant regimens and potentially allow personalisation perioperative therapies. Nevertheless, up to 30% of tissue samples fail to establish organoids in vivo, although the reasons for this are unclear. This study interrogates the clinical characteristics of biopsies that establish or fail to establish PDOs. Methods Esophageal adenocarcinomas from locally advanced (LA) or metastatic (M) disease were biopsied at initial endoscopy, following surgical resection, or image-guided biopsy of metastatic disease (n=75 total). Following a standardised protocol, cells were dissociated and seeded for organoid culture. Organoids are considered ‘established’ after passage 4. Samples were discarded if there was obvious contamination (n=9) or if there were no obvious tumour cells (n=6). Clinicopathologic variables for successfully established or unsuccessfully established were compared for both LA and M disease. Statistical analysis was performed to interrogate characteristics associated with successfully establishing organoids. Results There were significant differences between successful and unsuccessful organoids. Organoids were more likely to be established from endoscopic biopsies (p=0.002). Samples pre-treated with neoadjuvant chemoradiation or chemotherapy were less likely to grow (p=0.02). Samples from poorly differentiated tumours were less likely to grow compared with well and moderately differentiated tumours (p=0.05). Tumour stage, lymphovascular invasion, perineural invasion, MSI status and HER-2 status were not associated with growth characteristics. Conclusion Growth characteristics of esophageal adenocarcinoma PDOs correlate with clinicopathologic characteristics. Samples from patients who have undergone chemotherapy or chemoradiation, and poorly differentiated tumours are less likely to establish in vitro models. Understanding the altered in vitro growth patterns is key to translating use of organoids into clinical practice.

Clinical and translational mode of single-cell measurements: An artificial intelligent single-cell.

With rapid development and mature of single-cell measurements, single-cell biology and pathology become an emerging discipline to understand the disease. However, it is important to address concerns raised by clinicians as to how to apply single-cell measurements for clinical practice, translate the signals of single-cell systems biology into determination of clinical phenotype, and predict patient response to therapies. The present Perspective proposes a new system coined as the clinical artificial intelligent single-cell (caiSC) with the dynamic generator of clinical single-cell informatics, artificial intelligent analyzers, molecular multimodal reference boxes, clinical inputs and outs, and AI-based computerization. This system provides reliable and rapid information for impacting clinical diagnoses, monitoring, and prediction of the disease at the single-cell level. The caiSC represents an important step and milestone to translate the single-cell measurement into clinical application, assist clinicians' decision-making, and improve the quality of medical services. There is increasing evidence to support the possibility of the caiSC proposal, since the corresponding biotechnologies associated with caiSCs are rapidly developed. Therefore, we call the special attention and efforts from various scientists and clinicians on the caiSCs and believe that the appearance of the caiSCs can shed light on the future of clinical molecular medicine.

Distinct Gene Expression Patterns Identify Patients who Relapse After Neoadjuvant Pembrolizumab and Radical Cystectomy in the PURE-01 Study

PurposeThe PURE-01 clinical trial reported the use of neoadjuvant treatment with pembrolizumab prior to radical cystectomy (RC) in patients with muscle-invasive bladder. Specific molecular subtypes and immune signatures were reported to be associated with a favorable survival. However, reports on the detailed tumor biology of patients relapsing after neoadjuvant pembrolizumab are lacking. Materials and MethodsMicroarray data from transurethral resection of the bladder tumor (TURBT; n = 102) and matched RC (N = 25) tissue from patients in PURE-01 who experience a disease relapse were analyzed, with gene expression signatures and molecular subtypes. The Kaplan–Meier method was used to estimate differences in patient outcomes. Immune-signatures were split by median for survival analysis. All significance testing used a two-sided t-test at a threshold of P < .05. ResultsThe study cohort consisted of 102 patients, of whom N = 19 (19%) experienced relapse. Molecular subtyping revealed that neuroendocrine-like tumors had the worst outcomes, while tumors classified as Claudin-low did show only one recurrence event. Differential gene expression analysis identified genes associated with relapse, including KRT20, H19, and immune-associated genes such as CXCL9 and CXCL11. ConclusionThis study provides a detailed characterization of patients who relapsed after neoadjuvant pembrolizumab and RC and identifies distinct gene expression patterns associated with relapse. These findings may have implications for predicting patient response and guiding treatment decisions in the neoadjuvant setting.

Transforming Drug Discovery: The Impact of Artificial Intelligence and Machine Learning from Initial Screening to Clinical Trials

Abstract: The landscape of drug discovery is undergoing a profound transformation driven by the integration of Artificial Intelligence (AI) and Machine Learning (ML) technologies. This paper explores the impact of these advancements on various stages of the drug discovery process, from initial screening to clinical trials. Initially, AI and ML are revolutionizing the early stages of drug discovery by enhancing the efficiency of target identification and lead compound screening. Traditional methods, which often rely on labor-intensive processes and high costs, are being supplemented with AI-driven algorithms that analyze vast datasets to identify potential drug targets and predict the biological activity of compounds with unprecedented accuracy. In the drug design and optimization phase, ML models facilitate the prediction of drug interactions and side effects, thus accelerating the development of safer and more effective therapeutics. Advanced simulations and predictive models reduce the reliance on experimental trials, thereby streamlining the development pipeline. The clinical trials phase also benefits significantly from AI and ML. These technologies improve patient stratification by identifying suitable candidates based on genetic and clinical data, optimizing trial designs, and predicting patient responses to treatment. This not only enhances the efficiency of clinical trials but also increases the likelihood of successful outcome. Findings Our research highlights the transformative impact of Artificial Intelligence (AI) and Machine Learning (ML) on the drug discovery and development process, particularly in enhancing efficiency and precision from initial screening to clinical trials. The integration of AI/ML has shown significant advancements in early-stage drug discovery, where data-driven algorithms enable rapid identification of potential drug candidates, reducing reliance on traditional, labor-intensive methods. In the drug design and optimization phase, AI-driven predictive models have streamlined the process, minimizing the need for extensive physical testing by accurately simulating drug interactions and predicting possible side effects. Additionally, AI and ML are revolutionizing clinical trials by optimizing trial design, improving patient recruitment and retention, and enhancing real-time data monitoring, leading to faster and more reliable trial outcomes. These technologies also support personalized medicine approaches and have proven essential in reducing both the time and cost associated with bringing new therapies to market. Overall, our findings underscore the critical role of AI and ML in reshaping the pharmaceutical landscape, making drug development faster, more cost-effective, and ultimately, more successful in delivering effective treatments to patients.

Computational Methods for Predicting Key Interactions in T Cell-Mediated Adaptive Immunity.

The adaptive immune system recognizes pathogen- and cancer-specific features and is endowed with memory, enabling it to respond quickly and efficiently to repeated encounters with the same antigens. T cells play a central role in the adaptive immune system by directly targeting intracellular pathogens and helping to activate B cells to secrete antibodies. Several fundamental protein interactions-including those between major histocompatibility complex (MHC) proteins and antigen-derived peptides as well as between T cell receptors and peptide-MHC complexes-underlie the ability of T cells to recognize antigens with great precision. Computational approaches to predict these interactions are increasingly being used for medically relevant applications, including vaccine design and prediction of patient response to cancer immunotherapies. We provide computational researchers with an accessible introduction to the adaptive immune system, review computational approaches to predict the key protein interactions underlying T cell-mediated adaptive immunity, and highlight remaining challenges.

Trajectory modeling and response prediction in transcranial magnetic stimulation for depression

BackgroundRepetitive transcranial magnetic stimulation (rTMS) therapy could be improved by more accurate and earlier prediction of response. Latent class mixture (LCMM) and non-linear mixed effects (NLME) modeling have been applied to model the trajectories of antidepressant response (or non-response) to TMS, but it is not known whether such models are useful in predicting clinically meaningful change in symptom severity, i.e. categorical (non)response as opposed to continuous scores. MethodsWe compared LCMM and NLME approaches to model the antidepressant response to TMS in a naturalistic sample of 238 patients receiving rTMS for treatment resistant depression, across multiple coils and protocols. We then compared the predictive power of those models. ResultsLCMM trajectories were influenced largely by baseline symptom severity, but baseline symptoms provided little predictive power for later antidepressant response. Rather, the optimal LCMM model was a nonlinear two-class model that accounted for baseline symptoms. This model accurately predicted patient response at 4 weeks of treatment (AUC=0.70, 95 % CI=[0.52–0.87]), but not before. NLME offered slightly improved predictive performance at 4 weeks of treatment (AUC=0.76, 95 % CI=[0.58–0.94], but likewise, not before. ConclusionsIn showing the predictive validity of these approaches to model response trajectories to rTMS, we provided preliminary evidence that trajectory modeling could be used to guide future treatment decisions.

Baseline tumour vessel perfusion as a non-invasive predictive biomarker for immune checkpoint therapy in non-small-cell lung cancer

ObjectiveCurrent biomarkers for predicting immunotherapy response in non-small-cell lung cancer (NSCLC) are derived from invasive procedures with limited predictive accuracy. Thus, identifying a non-invasive predictive biomarker would improve patient stratification...

Intratumoral habitat radiomics based on magnetic resonance imaging for preoperative prediction treatment response to neoadjuvant chemotherapy in nasopharyngeal carcinoma.

The aim of this study is to determine intratumoral habitat regions from multi-sequences magnetic resonance imaging (MRI) and to assess the value of those regions for prediction of patient response to neoadjuvant chemotherapy (NAC) in nasopharyngeal carcinoma (NPC). Two hundred and ninety seven patients with NPC were enrolled. Multi-sequences MRI data were used to outline three-dimensional volumes of interest (VOI) of the whole tumor. The original imaging data were divided into two groups, which were resampled to an isotropic resolution of 1 × 1 × 1mm3 (group_1mm) and 3 × 3 × 3mm3 (group_3mm). Nineteen radiomics features were computed for each voxel of three sequences in group_3mm, within the tumor region to extract local information. Then, k-means clustering was implemented to segment the whole tumor regions in two groups. After radiomics features were extracted and dimension reduction, habitat models were built using Multi-Layer Perceptron (MLP) algorithm. Only T stage was included as the clinical model. The habitat3mm model, which included 10 radiomics features, achieved AUCs of 0.752 and 0.724 in the training and validation cohorts, respectively. Given the slightly better outcome of habitat3mm model, nomogram was developed in combination with habitat3mm model and T stage with the AUC of 0.749 and 0.738 in the training and validation cohorts. The decision curve analysis provides further evidence of the nomogram's clinical practicality. A nomogram based on intratumoral habitat predicts the efficacy of NAC in NPC patients, offering the potential to improve both the treatment plan and patient outcomes.

A bibliometric analysis of drug resistance in immunotherapy for breast cancer: trends, themes, and research focus.

While breast cancer treatments have advanced significantly nowadays, yet metastatic, especially triple-negative breast cancer (TNBC), remains challenging with low survival. Cancer immunotherapy, a promising approach for HER2-positive and TNBC, still faces resistance hurdles. Recently, numerous studies have set their sights on the resistance of immunotherapy for breast cancer. Our study provides a thorough comprehension of the current research landscape, hotspots, and emerging breakthroughs in this critical area through a meticulous bibliometric analysis. As of March 26, 2024, a total of 1341 articles on immunology resistance in breast cancer have been gathered from Web of Science Core Collection, including 765 articles and 576 reviews. Bibliometrix, CiteSpace and VOSviewer software were utilized to examine publications and citations per year, prolific countries, contributive institutions, high-level journals and scholars, as well as highly cited articles, references and keywords. The research of immunotherapy resistance in breast cancer has witnessed a remarkable surge over the past seven years. The United States and China have made significant contributions, with Harvard Medical School being the most prolific institution and actively engaging in collaborations. The most contributive author is Curigliano, G from the European Institute of Oncology in Italy, while Wucherpfennig, K. W. from the Dana-Farber Cancer Institute in the USA, had the highest citations. Journals highly productive primarily focus on clinical, immunology and oncology research. Common keywords include "resistance", "expression", "tumor microenvironment", "cancer", "T cell", "therapy", "chemotherapy" and "cell". Current research endeavors to unravel the mechanisms of immune resistance in breast cancer through theintegration of bioinformatics, basic experiments, and clinical trials. Efforts are underway to develop strategies that improve the effectiveness of immunotherapy, including the exploration of combination therapies and advancements in drug delivery systems. Additionally, there is a strong focus on identifying novel biomarkers that can predict patient response to immunology. This study will provide researchers with an up-to-date overview of the present knowledge in drug resistance of immunology for breast cancer, serving as a valuable resource for informed decision-making and further research on innovative approaches to address immunotherapy resistance.

The Spatial Structure of the Tumor Immune Microenvironment Can Explain and Predict Patient Response in High-Grade Serous Carcinoma.

Ovarian cancer is the deadliest gynecologic malignancy, and therapeutic options and mortality rates over the last three decades have largely not changed. Recent studies indicate that the composition of the tumor immune microenvironment (TIME) influences patient outcomes. To improve spatial understanding of the TIME, we performed multiplexed ion beam imaging on 83 human high-grade serous carcinoma tumor samples, identifying approximately 160,000 cells across 23 cell types. From the 77 of these samples that met inclusion criteria, we generated composition features based on cell type proportions, spatial features based on the distances between cell types, and spatial network features representing cell interactions and cell clustering patterns, which we linked to traditional clinical and IHC variables and patient overall survival (OS) and progression-free survival (PFS) outcomes. Among these features, we found several significant univariate correlations, including B-cell contact with M1 macrophages (OS HR = 0.696; P = 0.011; PFS HR = 0.734; P = 0.039). We then used high-dimensional random forest models to evaluate out-of-sample predictive performance for OS and PFS outcomes and to derive relative feature importance scores for each feature. The top model for predicting low or high PFS used TIME composition and spatial features and achieved an average AUC score of 0.71. The results demonstrate the importance of spatial structure in understanding how the TIME contributes to treatment outcomes. Furthermore, the present study provides a generalizable roadmap for spatial analyses of the TIME in ovarian cancer research.

IRnet: Immunotherapy response prediction using pathway knowledge-informed graph neural network

IntroductionImmune checkpoint inhibitors (ICIs) are potent and precise therapies for various cancer types, significantly improving survival rates in patients who respond positively to them. However, only a minority of patients benefit from ICI treatments. ObjectivesIdentifying ICI responders before treatment could greatly conserve medical resources, minimize potential drug side effects, and expedite the search for alternative therapies. Our goal is to introduce a novel deep-learning method to predict ICI treatment responses in cancer patients. MethodsThe proposed deep-learning framework leverages graph neural network and biological pathway knowledge. We trained and tested our method using ICI-treated patients’ data from several clinical trials covering melanoma, gastric cancer, and bladder cancer. ResultsOur results demonstrate that this predictive model outperforms current state-of-the-art methods and tumor microenvironment-based predictors. Additionally, the model quantifies the importance of pathways, pathway interactions, and genes in its predictions. A web server for IRnet has been developed and deployed, providing broad accessibility to users at https://irnet.missouri.edu. ConclusionIRnet is a competitive tool for predicting patient responses to immunotherapy, specifically ICIs. Its interpretability also offers valuable insights into the mechanisms underlying ICI treatments.

Clinical and translational mode of single‐cell measurements: An artificial intelligent single cell

AbstractWith rapid development and mature of single‐cell measurements, single‐cell biology and pathology become an emerging discipline to understand the disease. However, it is important to address concerns raised by clinicians as to how to apply single‐cell measurements for clinical practice, translate the signals of single‐cell systems biology into determination of clinical phenotype, and predict patient response to therapies. The present Perspective proposes a new system coined as the clinical artificial intelligent single‐cell (caiSC) with the dynamic generator of clinical single‐cell informatics, artificial intelligent analysers, molecular multimodal reference boxes, clinical inputs and outs, and AI‐based computerization. This system provides reliable and rapid information for impacting clinical diagnoses, monitoring, and prediction of the disease at the single‐cell level. The caiSC represents an important step and milestone to translate the single‐cell measurement into clinical application, assit clinicians’ decision‐making, and improve the quality of medical services. There is increasing evidence to support the possibility of the caiSC proposal, since the corresponding biotechnologies associated with caiSCs are rapidly developed. Therefore, we call the special attention and efforts from various scientists and clinicians on the caiSCs and believe that the appearance of the caiSCs can shed light on the future of clinical molecular medicine.

Pharmacogenomic Polygenic Model of Clopidogrel Predicts Recurrent Ischemic Events in Chinese Patients With Coronary Artery Disease

Patients with coronary artery disease (CAD) need to take antiplatelet drugs regularly in order to prevent thrombosis; however, there is existing inter-individual variability in drug response. Pharmacogenomic studies indicate that drug response may also be influenced by genetic variants, and multiple genetic variants may work together. We assumed that patients carrying more risk alleles might have a worse clopidogrel drug response and that a polygenic model integrated different single variants might have the potential to explain clopidogrel drug response variability better. We aimed to investigate whether the polygenic model could be used to predict clopidogrel drug response. A total of 935 CAD patients were enrolled in the study. We investigated the association between 19 clopidogrel-related single-nucleotide polymorphisms (SNPs) and the incidence of recurrent ischemic events. Additionally, a polygenic model was constructed to assess the risk of ischemic events. There were only 2 SNPs of CYP2C8 gene (rs1934980 and rs17110453) that were nominally associated with incidence of recurrent ischemic events. We constructed a polygenic model integrated with 6 clopidogrel-related SNPs. When compared with patients carrying 6 or fewer risk alleles, patients with 7 or more risk alleles had a higher risk of ischemic events (hazard ratio = 1.87; P = 0.04). The polygenetic model may be useful for clopidogrel drug response prediction in patients with CAD.

Eosinophils, Eosinophilic Gastrointestinal Diseases, and Inflammatory Bowel Disease: A Critical Review.

Inflammatory bowel disease (IBD) and eosinophilic gastrointestinal diseases (EGIDs) are complex, multifactorial chronic inflammatory disorders affecting the gastrointestinal tract. Their epidemiology, particularly for eosinophilic esophagitis (EoE), is increasing worldwide, with a rise in the co-diagnosis of IBD and EGIDs. Both disorders share common risk factors, such as early exposure to antibiotics or specific dietary habits. Moreover, from a molecular perspective, eosinophilic infiltration is crucial in the diagnosis of eosinophilic disorders, and it also plays a pivotal role in IBD histological diagnosis. Indeed, recent evidence highlights the significant role of eosinophils in the health of the intestinal mucosal barrier and as mediators between innate and acquired immunity, even indicating a potential role in IBD pathogenesis. This narrative review aims to summarize the current evidence regarding the common clinical and molecular aspects of EGIDs and IBD and the current state of knowledge regarding overlap conditions and their pathogenesis. Pubmed was searched until May 2023 to assess relevant studies describing the epidemiology, pathophysiology, and therapy of EGIDs in IBD. The immune pathways and mechanisms underlying both EGIDs and IBD remain partially known. An improved understanding of the role of eosinophils in overlapping conditions could lead to enhanced diagnostic precision, the development of more effective future therapeutic strategies, and a more accurate prediction of patient response. Consequently, the identification of red flags indicative of an eosinophilic disorder in IBD patients is of paramount importance and must be evaluated on a case-by-case basis.

GWO+RuleFit: rule-based explainable machine-learning combined with heuristics to predict mid-treatment FDG PET response to chemoradiation for locally advanced non-small cell lung cancer.

Objective.Vital rules learned from fluorodeoxyglucose positron emission tomography (FDG-PET) radiomics of tumor subregional response can provide clinical decision support for precise treatment adaptation. We combined a rule-based machine learning (ML) model (RuleFit) with a heuristic algorithm (gray wolf optimizer, GWO) for mid-chemoradiation FDG-PET response prediction in patients with locally advanced non-small cell lung cancer.Approach.Tumors subregions were identified using K-means clustering. GWO+RuleFit consists of three main parts: (i) a random forest is constructed based on conventional features or radiomic features extracted from tumor regions or subregions in FDG-PET images, from which the initial rules are generated; (ii) GWO is used for iterative rule selection; (iii) the selected rules are fit to a linear model to make predictions about the target variable. Two target variables were considered: a binary response measure (ΔSUVmean ⩾ 20% decline) for classification and a continuous response measure (ΔSUVmean) for regression. GWO+RuleFit was benchmarked against common ML algorithms and RuleFit, with leave-one-out cross-validated performance evaluated by the area under the receiver operating characteristic curve (AUC) in classification and root-mean-square error (RMSE) in regression.Main results.GWO+RuleFit selected 15 rules from the radiomic feature dataset of 23 patients. For treatment response classification, GWO+RuleFit attained numerically better cross-validated performance than RuleFit across tumor regions and sets of features (AUC: 0.58-0.86 vs. 0.52-0.78,p= 0.170-0.925). GWO+Rulefit also had the best or second-best performance numerically compared to all other algorithms for all conditions. For treatment response regression prediction, GWO+RuleFit (RMSE: 0.162-0.192) performed better numerically for low-dimensional models (p= 0.097-0.614) and significantly better for high-dimensional models across all tumor regions except one (RMSE: 0.189-0.219,p< 0.004).Significance. The GWO+RuleFit selected rules were interpretable, highlighting distinct radiomic phenotypes that modulated treatment response. GWO+Rulefit achieved parsimonious models while maintaining utility for treatment response prediction, which can aid clinical decisions for patient risk stratification, treatment selection, and biologically driven adaptation. Clinical trial: NCT02773238.

Plasma Metabolome Signatures to Predict Responsiveness to Neoadjuvant Chemotherapy in Breast Cancer.

Neoadjuvant chemotherapy (NACT) has arisen as a treatment option for breast cancer (BC). However, the response to NACT is still unpredictable and dependent on cancer subtype. Metabolomics is a tool for predicting biomarkers and chemotherapy response. We used plasma to verify metabolomic alterations in BC before NACT, relating to clinical data. Liquid chromatography coupled to mass spectrometry (LC-MS) was performed on pre-NACT plasma from patients with BC (n = 75). After data filtering, an SVM model for classification was built and validated with 75%/25% of the data, respectively. The model composed of 19 identified metabolites effectively predicted NACT response for training/validation sets with high sensitivity (95.4%/93.3%), specificity (91.6%/100.0%), and accuracy (94.6%/94.7%). In both sets, the panel correctly classified 95% of resistant and 94% of sensitive females. Most compounds identified by the model were lipids and amino acids and revealed pathway alterations related to chemoresistance. We developed a model for predicting patient response to NACT. These metabolite panels allow clinical gain by building precision medicine strategies based on tumor stratification.

Morinda officinalis oligosaccharides modulate the default-mode network homogeneity in major depressive disorder at rest

BackgroundWhile prior studies have explored the efficacy of Morinda officinalis oligosaccharides (MOs) as a treatment for patients with major depressive disorder (MDD), the mechanistic basis for the effects of MOs on brain function or the default-mode network (DMN) has yet to be characterized. The objective of this was to examine the effects of MOs treatment on functional connectivity in different regions of the DMN. MethodsIn total, 27 MDD patients and 29 healthy control subjects (HCs) underwent resting-state functional magnetic resonance imaging. The patients were then treated with MOs for 8 weeks, and scanning was performed at baseline and the end of the 8-week treatment period. Changes in DMN homogeneity associated with MOs treatment were assessed using network homogeneity (NH) analyses of the imaging data, and pattern classification approaches were employed to determine whether abnormal baseline NH deficits could differentiate between MDD patients and controls. The ability of NH abnormalities to predict patient responses to MOs treatment was also evaluated. ResultsRelative to HCs, patients exhibited a baseline reduction in NH values in the right precuneus (PCu). At the end of the 8-week treatment period, the MDD patients showed reduced and increased NH values in the right PCu and left superior medial frontal gyrus (SMFG), respectively. Compared to these patients at baseline, the 8-week MOs treatment was associated with reduced NH values in the right angular gyrus and increased NH values in the left middle temporal gyrus and the right PCu. Support vector machine (SVM) analyses revealed that NH abnormalities in the right PCu and left SMFG were the most accurate (87.50%) for differentiating between MDD patients and HCs. Conclusion: These results indicated that MOs treatment could alter default-mode NH in patients with MDD. The results provide a foundation for elucidation of the effects of MOs on brain function and suggest that the distinctive NH patterns observed in this study may be useful as imaging biomarkers for distinguishing between patients with MDD and healthy subjects.