Histopathological imaging features- versus molecular measurements-based cancer prognosis modeling.
Ontology highlight
ABSTRACT:
SUBMITTER: Zhang S
PROVIDER: S-EPMC7490375 | biostudies-literature | 2020 Sep
REPOSITORIES: biostudies-literature
Publications
Histopathological imaging features- versus molecular measurements-based cancer prognosis modeling.
Zhang Sanguo S Fan Yu Y Zhong Tingyan T Ma Shuangge S
Scientific reports 20200914 1
For lung and many other cancers, prognosis is essentially important, and extensive modeling has been carried out. Cancer is a genetic disease. In the past 2 decades, diverse molecular data (such as gene expressions and DNA mutations) have been analyzed in prognosis modeling. More recently, histopathological imaging data, which is a "byproduct" of biopsy, has been suggested as informative for prognosis. In this article, with the TCGA LUAD and LUSC data, we examine and directly compare modeling lu ...[more]
Similar Datasets
Project description:Liver transplantation is the solution for acute or chronic liver failure. There is an unmet demand for liver transplantation, as not all donated organs are transplanted. The clinical selection criteria for donated livers depends on histopathological evaluation and liver function tests. A fraction of the donor livers are rejected for transplantation due to the variability in applying the selection criteria. We applied an integrated transcriptomics and histopathological approach to characterize the donor livers evaluated for transplantation. We obtained liver biopsies from deceased donors during organ collection (n=10 accepted and n= 21 rejected for transplantation). We performed RNA sequencing to characterize the global gene expression profiles. We assessed steatosis, fibrosis, and necrosis via manual histopathological evaluation and a custom artificial intelligence-based image analysis. We identified two transcriptomically distinct subsets in the rejected donor liver group (termed rejected-1 and rejected-2), where the rejected-2 subset had a near complete transcriptomic overlap with the accepted livers, suggesting the potential for acceptability from a molecular standpoint. The liver metabolic functional genes were similarly upregulated in the accepted and rejected-2 groups compared to the rejected-1 subset. Expression levels of several extracellular matrix genes were similarly low in the accepted and rejected-2 groups compared to the rejected-1 subset. Filtering the rejected-2 subset based on histopathological evaluation identified the cases with borderline scores and extensive molecular overlap with the accepted group. Our integrated molecular and histopathological approach identified a subset of rejected donor livers that may be suitable for transplantation, thereby expanding the pool of transplantable livers.
2024-04-18 | GSE243887 | GEO
Project description:PurposeTo establish a pathomic model using histopathological image features for predicting indoleamine 2,3-dioxygenase 1 (IDO1) status and its relationship with overall survival (OS) in breast cancer.MethodsA pathomic model was constructed using machine learning and histopathological images obtained from The Cancer Genome Atlas database to predict IDO1 expression. The model performance was evaluated based on the area under the curve, calibration curve, and decision curve analysis (DCA). Prediction scores (PSes) were generated from the model and applied to divide the patients into two groups. Survival outcomes, gene set enrichment, immune microenvironment, and tumor mutations were assessed between the two groups.ResultsSurvival analysis followed by multivariate correction revealed that high IDO1 is a protective factor for OS. Further, the model was calibrated, and it exhibited good discrimination. Additionally, the DCA showed that the proposed model provided a good clinical net benefit. The Kaplan-Meier analysis revealed a positive correlation between high PS and improved OS. Univariate and multivariate Cox regression analyses demonstrated that PS is an independent protective factor for OS. Moreover, differentially expressed genes were enriched in various essential biological processes, including extracellular matrix receptor interaction, angiogenesis, transforming growth factor β signaling, epithelial mesenchymal transition, cell junction, tryptophan metabolism, and heme metabolic processes. PS was positively correlated with M1 macrophages, CD8 + T cells, T follicular helper cells, and tumor mutational burden.ConclusionThese results indicate the potential ability of the proposed pathomic model to predict IDO1 status and the OS of breast cancer patients.
| S-EPMC11230954 | biostudies-literature
Project description:Cancer prognosis is of essential interest, and extensive research has been conducted searching for biomarkers with prognostic power. Recent studies have shown that both omics profiles and histopathological imaging features have prognostic power. There are also studies exploring integrating the two types of measurements for prognosis modeling. However, there is a lack of study rigorously examining whether omics measurements have independent prognostic power conditional on histopathological imaging features, and vice versa. In this article, we adopt a rigorous statistical testing framework and test whether an individual gene expression measurement can improve prognosis modeling conditional on high-dimensional imaging features, and a parallel analysis is conducted reversing the roles of gene expressions and imaging features. In the analysis of The Cancer Genome Atlas (TCGA) lung adenocarcinoma and liver hepatocellular carcinoma data, it is found that multiple individual genes, conditional on imaging features, can lead to significant improvement in prognosis modeling; however, individual imaging features, conditional on gene expressions, only offer limited prognostic power. Being among the first to examine the independent prognostic power, this study may assist better understanding the "connectedness" between omics profiles and histopathological imaging features and provide important insights for data integration in cancer modeling.
| S-EPMC6468814 | biostudies-other
Project description:Pre-clinical bone cancer pain models mimicking the human condition are required to respond to clinical realities. Breast or prostate cancer patients coping with bone metastases experience intractable pain, which affects their quality of life. Advanced monitoring is thus required to clarify bone cancer pain mechanisms and refine treatments. In our model of rat femoral mammary carcinoma MRMT-1 cell implantation, pain onset and tumor growth were monitored for 21 days. The surgical procedure performed without arthrotomy allowed recording of incidental pain in free-moving rats. Along with the gradual development of mechanical allodynia and hyperalgesia, behavioral signs of ambulatory pain were detected at day 14 by using a dynamic weight-bearing apparatus. Osteopenia was revealed from day 14 concomitantly with disorganization of the trabecular architecture (µCT). Bone metastases were visualized as early as day 8 by MRI (T(1)-Gd-DTPA) before pain detection. PET (Na(18)F) co-registration revealed intra-osseous activity, as determined by anatomical superimposition over MRI in accordance with osteoclastic hyperactivity (TRAP staining). Pain and bone destruction were aggravated with time. Bone remodeling was accompanied by c-Fos (spinal) and ATF3 (DRG) neuronal activation, sustained by astrocyte (GFAP) and microglia (Iba1) reactivity in lumbar spinal cord. Our animal model demonstrates the importance of simultaneously recording pain and tumor progression and will allow us to better characterize therapeutic strategies in the future.
| S-EPMC2966439 | biostudies-other
Project description:BackgroundThis study aimed to construct a prognostic stratification system for gastric cancer (GC) using tumour invasion-related genes to more accurately predict the clinical prognosis of GC.MethodologyTumour invasion-related genes were downloaded from CancerSEA, and their expression data in the TCGA-STAD dataset were used to cluster samples via non-negative matrix factorisation (NMF). Differentially expressed genes (DEGs) between subtypes were identified using the limma package. KEGG pathway and GO functional enrichment analyses were conducted using the WebGestaltR package (v0.4.2). The immune scores of molecular subtypes were evaluated using the R package ESTIMATE, MCPcounter and the ssGSEA function of the GSVA package. Univariate, multivariate and lasso regression analyses of DEGs were performed using the coxph function of the survival package and the glmnet package to construct a RiskScore model. The robustness of the model was validated using internal and external datasets, and a nomogram was constructed based on the model.ResultsBased on 97 tumour invasion-related genes, 353 GC samples from TCGA were categorised into two subtypes, thereby indicating the presence of inter-subtype differences in prognosis. A total of 569 DEGs were identified between the two subtypes; of which, four genes were selected to construct the risk model. This four-gene signature was robust and exhibited stable predictive performance in different platform datasets (GSE26942 and GSE66229), indicating that the established model performed better than other existing models.ConclusionA prognostic stratification system based on a four-gene signature was developed with a desirable area under the curve in the training and independent validation sets. Therefore, the use of this system as a molecular diagnostic test is recommended to assess the prognostic risk of patients with GC.
| S-EPMC9203697 | biostudies-literature
Project description:The incidence of oropharyngeal squamous cell carcinoma (OPSCC) has been rapidly increasing. Disease stage and smoking history are often used in current clinical trials to select patients for deintensification therapy, but these features lack sufficient accuracy for predicting disease relapse. Our purpose was to develop an imaging signature to assess early response and predict outcomes of OPSCC. Methods: We retrospectively analyzed 162 OPSCC patients treated with concurrent chemoradiotherapy, equally divided into separate training and validation cohorts with similar clinical characteristics. A robust consensus clustering approach was used to spatially partition the primary tumor and involved lymph nodes into subregions (i.e., habitats) based on 18F-FDG PET and contrast CT imaging. We proposed quantitative image features to characterize the temporal volumetric change of the habitats and peritumoral/nodal tissue between baseline and midtreatment. The reproducibility of these features was evaluated. We developed an imaging signature to predict progression-free survival (PFS) by fitting an L1-regularized Cox regression model. Results: We identified 3 phenotypically distinct intratumoral habitats: metabolically active and heterogeneous, enhancing and heterogeneous, and metabolically inactive and homogeneous. The final Cox model consisted of 4 habitat evolution-based features. In both cohorts, this imaging signature significantly outperformed traditional imaging metrics, including midtreatment metabolic tumor volume for predicting PFS, with a C-index of 0.72 versus 0.67 (training) and 0.66 versus 0.56 (validation). The imaging signature stratified patients into high-risk versus low-risk groups with 2-y PFS rates of 59.1% versus 89.4% (hazard ratio, 4.4; 95% confidence interval, 1.4-13.4 [training]) and 61.4% versus 87.8% (hazard ratio, 4.6; 95% confidence interval, 1.7-12.1 [validation]). The imaging signature remained an independent predictor of PFS in multivariable analysis adjusting for stage, human papillomavirus status, and smoking history. Conclusion: The proposed imaging signature allows more accurate prediction of disease progression and, if prospectively validated, may refine OPSCC patient selection for risk-adaptive therapy.
| S-EPMC7067523 | biostudies-literature
Project description:BackgroundHistopathological image features offer a quantitative measurement of cellular morphology, and probably help for better diagnosis and prognosis in head and neck squamous cell carcinoma (HNSCC).MethodsWe first used histopathological image features and machine-learning algorithms to predict molecular features of 212 HNSCC patients from The Cancer Genome Atlas (TCGA). Next, we divided TCGA-HNSCC cohort into training set (n = 149) and test set (n = 63), and obtained tissue microarrays as an external validation set (n = 126). We identified the gene expression profile correlated to image features by bioinformatics analysis.ResultsHistopathological image features combined with random forest may predict five somatic mutations, transcriptional subtypes, and methylation subtypes, with area under curve (AUC) ranging from 0.828 to 0.968. The prediction model based on image features could predict overall survival, with 5-year AUC of 0.831, 0.782, and 0.751 in training, test, and validation sets. We next established an integrative prognostic model of image features and gene expressions, which obtained better performance in training set (5-year AUC = 0.860) and test set (5-year AUC = 0.826). According to histopathological transcriptomics risk score (HTRS) generated by the model, high-risk and low-risk patients had different survival in training set (HR = 4.09, p < 0.001) and test set (HR=3.08, p = 0.019). Multivariate analysis suggested that HTRS was an independent predictor in training set (HR = 5.17, p < 0.001). The nomogram combining HTRS and clinical factors had higher net benefit than conventional clinical evaluation.ConclusionsHistopathological image features provided a promising approach to predict mutations, molecular subtypes, and prognosis of HNSCC. The integration of image features and gene expression data had potential for improving prognosis prediction in HNSCC.
| S-EPMC8267162 | biostudies-literature
Project description:ObjectiveAging acts as a dominating risk factor for human cancers. Herein, we systematically dissected the features of transcriptional aging-relevant genes in gastric cancer from multiple perspectives.MethodsBased on the transcriptome profiling of prognostic aging-relevant genes, patients with gastric cancer in The Cancer Genome Atlas (TCGA) stomach adenocarcinoma (TCGA-STAD) cohort were clustered with a consensus clustering algorithm. Mutational landscape and chemotherapeutic responses were analyzed and immunological features (immunomodulators, immune checkpoint molecules, cancer immunity cycle, and tumor-infiltrating immune cells) were systematically evaluated across gastric cancer. Weighted gene co-expression network (WGCNA) was conducted for screening aging molecular phenotype-relevant genes, and key genes were identified with Molecular Complex Detection (MCODE) analyses. Expressions of key genes were examined in 20 paired tumors and controls with RT-qPCR and Western blotting. Proliferation and apoptosis were investigated in two gastric cancer cells under MYL9 deficiency.ResultsThree aging-based molecular phenotypes (namely, C1, C2, and C3) were conducted in gastric cancer. Phenotype C1 presented the most prominent survival advantage and highest mutational frequencies. Phenotype C2 indicated low responses to sorafenib and gefitinib, while C3 indicated low responses to vinorelbine and gemcitabine. Additionally, phenotype C2 was characterized by enhanced immune and stromal activation and an inflamed tumor microenvironment. Seven aging molecular phenotype-relevant key genes (ACTA2, CALD1, LMOD1, MYH11, MYL9, MYLK, and TAGLN) were identified, which were specifically upregulated in tumors and in relation to dismal prognosis. Among them, MYL9 deficiency reduced proliferation and enhanced apoptosis in gastric cancer cells.ConclusionCollectively, aging-based molecular subtypes may offer more individualized therapy recommendations and prognosis assessment for patients in distinct subtypes.
| S-EPMC8843835 | biostudies-literature
Project description:The repository is composed of 1224 images divided into two sets of images with two different resolutions. First set consists of 89 histopathological images with the normal epithelium of the oral cavity and 439 images of Oral Squamous Cell Carcinoma (OSCC) in 100x magnification. The second set consists of 201 images with the normal epithelium of the oral cavity and 495 histopathological images of OSCC in 400x magnification. The images were captured using a Leica ICC50 HD microscope from Hematoxyline and Eosin (H&E) stained tissue slides collected, prepared and catalogued by medical experts from 230 patients. A subset of 269 images from the second data set was used to detect OSCC based on textural features [1]. Histopathology plays a very important role in diagnosing a disease. It is the investigation of biological tissues to detect the presence of diseased cells in microscopic detail. It usually involves a biopsy. Till date biopsy is the gold-standard test to diagnose cancer. The biopsy slides are examined based on various cytological criteria under a microscope. Therefore, there is a high possibility of not retaining uniformity and ensuring reproducibility in outcomes [2, 3]. Computational diagnostic tools, on the other hand, facilitate objective judgments by making the use of the quantitative measure. This dataset can be utilized in establishing automated diagnostic tool using Artificial Intelligence approaches.
| S-EPMC6994517 | biostudies-literature
Project description:Cancer is heterogeneous, and for seemingly similar cancer patients, the associations between an outcome/phenotype and covariates can be different. To describe such differences, finite mixture of regression (FMR) and other modeling techniques have been developed. "Classic" FMR analysis has usually been based on clinical, demographic, and molecular variables. More recently, histopathological imaging data-which is a byproduct of biopsy and enjoys broader data availability and higher cost-effectiveness-has been increasingly used in cancer modeling, although it is noted that its application to cancer FMR analysis still remains limited. In this article, we further advance cancer FMR analysis based on histopathological imaging data. Significantly advancing from the existing analyses under heterogeneity and homogeneity, our goal is to simultaneously use two types of histopathological imaging features, which are extracted based on domain-specific biomedical knowledge and using automated signal processing software, respectively. A significant modeling/methodological advancement is that, to reflect the "increased resolution" of the second type of imaging features over the first type, we impose a hierarchy in the mixture structures. An effective and flexible Bayesian approach is proposed. Simulation shows its competitiveness over several alternatives. The TCGA lung cancer data is analyzed, and interesting heterogeneous structures different from using the alternatives are found. Overall, this study provides a new venue for FMR analysis for cancer and other complex diseases.
| S-EPMC8881390 | biostudies-literature