scholarly journals Predicting Endometrial Cancer Subtypes and Molecular Features from Histopathology Images Using Multi-resolution Deep Learning Models

2020 ◽  
Author(s):  
Runyu Hong ◽  
Wenke Liu ◽  
Deborah DeLair ◽  
Narges Razavian ◽  
David Fenyö
Keyword(s):  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Characteristic Curve ◽  
The Cancer Genome Atlas ◽  
Diagnostic Process ◽  
Histological Subtypes ◽  
Neural Network Models ◽  
Molecular Subtyping ◽  
Cancer Subtypes ◽  
Molecular Features

AbstractThe determination of endometrial carcinoma histological subtypes is a critical diagnostic process that directly affects patients’ prognosis and treatment options. Recently, molecular subtyping and mutation status are increasingly utilized in clinical practice as they offer better inform prognosis and offer the possibility of individualized therapies. Compared to the histopathological approach, however, the availability of molecular subtyping is limited as it can only be obtained by genomic sequencing, which may be cost prohibitive. Here, we implemented deep convolutional neural network models that predict not only the histological subtypes, but also molecular subtypes and 18 common gene mutations based on digitized H&E stained pathological images. Taking advantage of the multi-resolution nature of the whole slide images, we introduced a customized architecture, Panoptes, to integrate features of different magnification. The model was trained and evaluated with images from The Cancer Genome Atlas and Clinical Proteomic Tumor Analysis Consortium. Our models achieved an area under the receiver operating characteristic curve (AUROC) of 0.969 in predicting histological subtype and 0.934 to 0.958 in predicting the copy number high (CNV-H) molecular subtype. The prediction tasks of 4 mutations and microsatellite high (MSI-H) molecular subtype also achieved a high performance with AUROC ranging from 0.781 to 0.873. Panoptes showed a significantly better performance than InceptionResnet in most of these top predicted tasks by up to 18%. Feature extraction and visualization revealed that the model relied on human-interpretable patterns. Our results suggest that Panoptes can help pathologists determine molecular subtypes and mutations without sequencing, and our models are generalizable to independent datasets.

scholarly journals Highly connected, non-redundant microRNA functional control in breast cancer molecular subtypes

2021 ◽  
Vol 11 (4) ◽  
pp. 20200073 ◽  
Author(s):  
Guillermo de Anda-Jáuregui ◽  
Jesús Espinal-Enríquez ◽  
Enrique Hernández-Lemus
Keyword(s):  
Breast Cancer ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Expression Patterns ◽  
Network Biology ◽  
The Cancer Genome Atlas ◽  
Biological Functions ◽  
Cancer Subtypes ◽  
Potential Control ◽  
Functional Studies

Breast cancer is a complex, heterogeneous disease at the phenotypic and molecular level. In particular, the transcriptional regulatory programs are known to be significantly affected and such transcriptional alterations are able to capture some of the heterogeneity of the disease, leading to the emergence of breast cancer molecular subtypes. Recently, it has been found that network biology approaches to decipher such abnormal gene regulation programs, for instance by means of gene co-expression networks, have been able to recapitulate the differences between breast cancer subtypes providing elements to further understand their functional origins and consequences. Network biology approaches may be extended to include other co-expression patterns, like those found between genes and non-coding transcripts such as microRNAs (miRs). As is known, miRs play relevant roles in the establishment of normal and anomalous transcription processes. Commodore miRs (cdre-miRs) have been defined as miRs that, based on their connectivity and redundancy in co-expression networks, are potential control elements of biological functions. In this work, we reconstructed miR–gene co-expression networks for each breast cancer molecular subtype, from high throughput data in 424 samples from the Cancer Genome Atlas consortium. We identified cdre-miRs in three out of four molecular subtypes. We found that in each subtype, each cdre-miR was linked to a different set of associated genes, as well as a different set of associated biological functions. We used a systematic literature validation strategy, and identified that the associated biological functions to these cdre-miRs are hallmarks of cancer such as angiogenesis, cell adhesion, cell cycle and regulation of apoptosis. The relevance of such cdre-miRs as actionable molecular targets in breast cancer is still to be determined from functional studies.

scholarly journals DNA Methylation Based Molecular Subtypes Predict Prognosis in Breast Cancer Patients

2021 ◽  
Vol 28 ◽  
pp. 107327482098851
Author(s):  
Zeng-Hong Wu ◽  
Yun Tang ◽  
Yan Zhou
Keyword(s):  
Breast Cancer ◽  
Dna Methylation ◽  
Molecular Subtypes ◽  
Methylation Status ◽  
The Cancer Genome Atlas ◽  
Training Dataset ◽  
Consensus Clustering ◽  
Breast Cancer Patients ◽  
Cancer Subtypes ◽  
Novel Biomarkers

Background: Epigenetic changes are tightly linked to tumorigenesis development and malignant transformation’ However, DNA methylation occurs earlier and is constant during tumorigenesis. It plays an important role in controlling gene expression in cancer cells. Methods: In this study, we determining the prognostic value of molecular subtypes based on DNA methylation status in breast cancer samples obtained from The Cancer Genome Atlas database (TCGA). Results: Seven clusters and 204 corresponding promoter genes were identified based on consensus clustering using 166 CpG sites that significantly influenced survival outcomes. The overall survival (OS) analysis showed a significant prognostic difference among the 7 groups (p<0.05). Finally, a prognostic model was used to estimate the results of patients on the testing set based on the classification findings of a training dataset DNA methylation subgroups. Conclusions: The model was found to be important in the identification of novel biomarkers and could be of help to patients with different breast cancer subtypes when predicting prognosis, clinical diagnosis and management.

scholarly journals Novel cancer subtyping method based on patient-specific gene regulatory network

2021 ◽  
Author(s):  
Mai Adachi Nakazawa ◽  
Yoshinori Tamada ◽  
Yoshihisa Tanaka ◽  
Marie Ikeguchi ◽  
Kako Higashihara ◽  
...  
Keyword(s):  
Gene Networks ◽  
Regulatory Networks ◽  
The Cancer Genome Atlas ◽  
Patient Specific ◽  
Specific Gene ◽  
Omics Data ◽  
Cancer Subtypes ◽  
Molecular Systems ◽  
Molecular Features ◽  
Cancer Genome Atlas

The identification of cancer subtypes is important for the understanding of tumor heterogeneity. In recent years, numerous computational methods have been proposed for this problem based on the multi-omics data of patients. It is widely accepted that different cancer subtypes are induced by different molecular regulatory networks. However, only a few incorporate the differences between their molecular systems into the classification processes. In this study, we present a novel method to classify cancer subtypes based on patient-specific molecular systems. Our method quantifies patient-specific gene networks, which are estimated from their transcriptome data. By clustering their quantified networks, our method allows for cancer subtyping, taking into consideration the differences in the molecular systems of patients. Comprehensive analyses of The Cancer Genome Atlas (TCGA) datasets applied to our method confirmed that they were able to identify more clinically meaningful cancer subtypes than the existing subtypes and found that the identified subtypes comprised different molecular features. Our findings show that the proposed method, based on a simple classification using the patient-specific molecular systems, can identify cancer subtypes even with single omics data, which cannot otherwise be captured by existing methods using multi-omics data.

scholarly journals Differential methylation of G-protein coupled receptor signaling genes in gastrointestinal neuroendocrine tumors

2021 ◽  
Author(s):  
Seyoun Byun ◽  
Kajsa E. Affolter ◽  
Angela K. Snow ◽  
Karen Curtin ◽  
Austin R. Cannon ◽  
...  
Keyword(s):  
Small Intestine ◽  
Neuroendocrine Tumors ◽  
Patient Survival ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Treatment Strategies ◽  
Molecular Features ◽  
Differentially Methylated Genes ◽  
Chromosomal Changes ◽  
Approved Drugs

Abstract Neuroendocrine tumors (NETs) of the small intestine undergo large chromosomal and methylation changes. The objective of this study was to identify methylation differences in NETs and consider how the differentially methylated genes may impact patient survival. Whole genome methylation and chromosomal copy number varation (CNV) of NETs from the small intestine and appendix was measured Tumors were divided into three molecular subtypes according to CNV results: chromosome 18 loss (18LOH), MultiCNV, and NoCNV. Comparison of 18LOH tumors with MultiCNV and NoCNV tumors identified 901 differentially methylated genes which are over represented with sevenG-protein coupled receptor (GPCR) pathways and the gene encoding somatostatin (SST), a clinical target for NETs. Patient survival based on low versus high methylation in all samples identified four significant genes (p-<0.05) OR2S2, SMILR, RNU6-653P and AC010543.1. Within the 18LOH molecular subtype tumors, survival differences were identifiedin high versus low methylation of 24 genes. The most significant is TRHR (p < 0.01), a GPCR with multiple FDA-approved drugs. By separating NETs into different molecular subtypes based on chromosomal changes, we find that multiple GPCRs and their ligands appear to be regulated through methylation and correlated with survival. Opportunities for better treatment strategies based on molecular features exist for NETs.

scholarly journals Highly-connected, non-redundant microRNAs functional control in breast cancer molecular subtypes

2019 ◽  
Author(s):  
Guillermo de Anda-Jáuregui ◽  
Jesús Espinal-Enríquez ◽  
Enrique Hernández-Lemus
Keyword(s):  
Breast Cancer ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Expression Patterns ◽  
Network Biology ◽  
Biological Functions ◽  
Cancer Subtypes ◽  
Potential Control ◽  
Non Coding Rna ◽  
Functional Control

AbstractTranscriptional patterns are altered in breast cancer. These alterations capture the heterogeneity of breast cancer, leading to the emergence of molecular subtypes. Network biology approaches to study gene co-expression are able to capture the differences between breast cancer subtypes.Network biology approaches may be extended to include other co-expression patterns, like those found between genes and non-coding RNA: such as mi-croRNAs (miRs). Commodore miRs are microRNAs that, based on their connectivity and redundancy in co-expression networks, have been proposed as potential control elements of biological functions.In this work, we reconstructed miR-gene co-expression networks for each breast cancer molecular subtype. We identified Commodore miRs in three out of four molecular subtypes. We found that in each subtype, each cdre-miR had a different set of associated genes, as well as a different set of associated biological functions. We used a systematic literature validation strategy, and identified that the associated biological functions to these cdre-miRs are hallmarks of cancer.

Tumor sidedness and intrinsic subtypes in patients with stage II/III colon cancer: Analysis of NSABP C-07 (NRG Oncology).

2017 ◽  
Vol 35 (15_suppl) ◽  
pp. 3514-3514 ◽  
Author(s):  
S. Rim Kim ◽  
Nan Song ◽  
Greg Yothers ◽  
Patrick Gavin ◽  
Carmen Joseph Allegra ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Colon Cancer ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Therapy Response ◽  
Stage Ii ◽  
Intrinsic Subtypes ◽  
Treatment Benefit ◽  
Cancer Subtypes ◽  
Metastatic Setting

3514 Background: The predictive value of tumor sidedness in colorectal cancer is currently of interest especially in metastatic setting for anti-EGFR therapy response. We tested whether intrinsic molecular subtype classification predictive of treatment benefit in stage II/III colon cancer is an independent novel marker in association with tumor sidedness. Methods: All available cases included in the NSABP/NRG C-07 trial for which we had both gene expression data and anatomical data (n=1603) were used to determine the molecular subtypes using the following classifiers; the Colorectal Cancer Assigner (CRCA), the Colon Cancer Subtypes (CCS) and the Consensus Molecular Subtypes (CMS). Frequency of tumor sidedness in each subtype and recurrence-free survival were analyzed. Results: Intrinsic subtypes were differentially distributed in right- and left-colon tumors with the exception of the stem-like or CMS4 (mesenchymal) subtype (Table 1). Sidedness was not associated with prognosis (p=0.82, HR: 1.022 [CI: 0.851-1.227]) or prediction of patients with greater benefit from oxaliplatin when combined with 5-Fu+LV (interaction p=0.484). Conclusions: Although tumor sidedness is associated with distribution of intrinsic subtypes in stage II/III colon cancer, it is not predictive of survival benefit from oxaliplatin in C-07. Support: -180868, -180822, U24-CA196067; HI13C2162; PA DOH; Sanofi-Synthelabo Clinical trial information: NCT00004931. [Table: see text]

Muscle invasive bladder cancer (MIBC) demonstrates neoadjuvant cisplatin-based chemotherapy (NAC) related changes in molecular subtype and immune infiltration.

2019 ◽  
Vol 37 (7_suppl) ◽  
pp. 443-443
Author(s):  
Samuel Aaron Funt ◽  
Alexander Solovyov ◽  
Bishoy Morris Faltas ◽  
Gopa Iyer ◽  
Mariel Elena Boyd ◽  
...  
Keyword(s):  
Bladder Cancer ◽  
Bladder Tumor ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Expression Patterns ◽  
The Cancer Genome Atlas ◽  
Rna Expression ◽  
Cancer Therapies ◽  
Primary Bladder ◽  
Muscle Invasive

443 Background: Defining the role of MIBC molecular subtypes and immune expression in determining clinical outcomes is an area of active investigation. However, changes in these transcriptomic profiles pre- and post-NAC have not been well characterized. Methods: This retrospective study reviewed 53 pts with MIBC treated with NAC, of whom 12 pts without complete pathological response had both pre- and post-NAC samples of sufficient quality. Post-NAC staging was > = pT2 in 11 pts and pT1 in 1 pt. We performed RNA expression analysis of matched pre-NAC transurethral resection of bladder tumor specimens and post-treatment radical cystectomy primary bladder tumor specimens. We used a customized NanoString panel incorporating previously reported immune signatures (Ayers, JCI 2017; O’Donnell, ASCO 2017) and additional genes to assign basal ( CD14, CD44, PDGFC, KRT14, KRT5) and luminal ( GATA3, PPARG, SHH, CD24, FOXA1, WNT7B, ERBB2) molecular subtypes. Results: We first classified the bladder cancer cohort of The Cancer Genome Atlas into basal and luminal subtypes using the BASE47 signature (Damrauer, PNAS 2014) and the NanoString panel and there was good agreement (Rand Index = 0.72). We then assigned subtypes using the NanoString panel on matched pre- and post-NAC samples and found marked subtype shift (Table). We identified two robust clusters of samples according to immune expression with a 3-fold change of immune expression between them (FDR = 0.0008). We found that 4 pts switched from the low to the high cluster, while 2 switched from the high to the low cluster after NAC (Table). Conclusions: MIBC molecular subtype membership is dynamic and is influenced by NAC. NAC can induce both enhanced and suppressed immune activity. These findings have implications on future studies exploring the predictive value of RNA expression patterns for bladder cancer therapies as well as post-NAC immunotherapy. [Table: see text]

scholarly journals Differential methylation of G-protein coupled receptor signaling genes in gastrointestinal neuroendocrine tumors

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Seyoun Byun ◽  
Kajsa E. Affolter ◽  
Angela K. Snow ◽  
Karen Curtin ◽  
Austin R. Cannon ◽  
...  
Keyword(s):  
Small Intestine ◽  
G Protein ◽  
Neuroendocrine Tumors ◽  
Patient Survival ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
G Protein Coupled Receptor ◽  
Molecular Features ◽  
Differentially Methylated Genes ◽  
G Protein Coupled

AbstractNeuroendocrine tumors (NETs) of the small intestine undergo large chromosomal and methylation changes. The objective of this study was to identify methylation differences in NETs and consider how the differentially methylated genes may impact patient survival. Genome-wide methylation and chromosomal copy number variation (CNV) of NETs from the small intestine and appendix were measured. Tumors were divided into three molecular subtypes according to CNV results: chromosome 18 loss (18LOH), Multiple CNV, and No CNV. Comparison of 18LOH tumors with MultiCNV and NoCNV tumors identified 901 differentially methylated genes. Genes from the G-protein coupled receptor (GPCR) pathways are statistically overrepresented in the differentially methylated genes. One of the highlighted genes from the GPCR pathway is somatostatin (SST), a clinical target for NETs. Patient survival based on low versus high methylation in all samples identified four significant genes (p < 0.05) OR2S2, SMILR, RNU6-653P, and AC010543.1. Within the 18LOH molecular subtype tumors, survival differences were identified in high versus low methylation of 24 genes. The most significant is TRHR (p < 0.01), a GPCR with multiple FDA-approved drugs. By separating NETs into different molecular subtypes based on chromosomal changes, we find that multiple GPCRs and their ligands appear to be regulated through methylation and correlated with survival. These results suggest opportunities for better treatment strategies for NETs based on molecular features.

scholarly journals Computer-Aided Diagnosis Evaluation of the Correlation Between Magnetic Resonance Imaging With Molecular Subtypes in Breast Cancer

2021 ◽  
Vol 11 ◽  
Author(s):  
Wei Meng ◽  
Yunfeng Sun ◽  
Haibin Qian ◽  
Xiaodan Chen ◽  
Qiujie Yu ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Mr Imaging ◽  
Clinical Features ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
Texture Features ◽  
Breast Cancer Subtypes ◽  
Cancer Subtypes ◽  
Computer Aided ◽  
Aided Diagnosis

BackgroundThere is a demand for additional alternative methods that can allow the differentiation of the breast tumor into molecular subtypes precisely and conveniently.PurposeThe present study aimed to determine suitable optimal classifiers and investigate the general applicability of computer-aided diagnosis (CAD) to associate between the breast cancer molecular subtype and the extracted MR imaging features.MethodsWe analyzed a total of 264 patients (mean age: 47.9 ± 9.7 years; range: 19–81 years) with 264 masses (mean size: 28.6 ± 15.86 mm; range: 5–91 mm) using a Unet model and Gradient Tree Boosting for segmentation and classification.ResultsThe tumors were segmented clearly by the Unet model automatically. All the extracted features which including the shape features,the texture features of the tumors and the clinical features were input into the classifiers for classification, and the results showed that the GTB classifier is superior to other classifiers, which achieved F1-Score 0.72, AUC 0.81 and score 0.71. Analyzed the different features combinations, we founded that the texture features associated with the clinical features are the optimal features to different the breast cancer subtypes.ConclusionCAD is feasible to differentiate the breast cancer subtypes, automatical segmentation were feasible by Unet model and the extracted texture features from breast MR imaging with the clinical features can be used to help differentiating the molecular subtype. Moreover, in the clinical features, BPE and age characteristics have the best potential for subtype.

scholarly journals Identification and Validation of Immune Molecular Subtypes in Pancreatic Ductal Adenocarcinoma: Implications for Prognosis and Immunotherapy

2021 ◽  
Vol 12 ◽  
Author(s):  
Ruiyu Li ◽  
Yangzhige He ◽  
Hui Zhang ◽  
Jing Wang ◽  
Xiaoding Liu ◽  
...  
Keyword(s):  
Overall Survival ◽  
Prognostic Factor ◽  
Pancreatic Ductal Adenocarcinoma ◽  
Immune Cell ◽  
Molecular Subtypes ◽  
Molecular Subtype ◽  
The Cancer Genome Atlas ◽  
Ductal Adenocarcinoma ◽  
Expression Data ◽  
Favorable Prognostic Factor

BackgroundPancreatic ductal adenocarcinoma (PDAC) remains treatment refractory. Immunotherapy has achieved success in the treatment of multiple malignancies. However, the efficacy of immunotherapy in PDAC is limited by a lack of promising biomarkers. In this research, we aimed to identify robust immune molecular subtypes of PDAC to facilitate prognosis prediction and patient selection for immunotherapy.MethodsA training cohort of 149 PDAC samples from The Cancer Genome Atlas (TCGA) with mRNA expression data was analyzed. By means of non-negative matrix factorization (NMF), we virtually dissected the immune-related signals from bulk gene expression data. Detailed immunogenomic and survival analyses of the immune molecular subtypes were conducted to determine their biological and clinical relevance. Validation was performed in five independent datasets on a total of 615 samples.ResultsApproximately 31% of PDAC samples (46/149) had higher immune cell infiltration, more active immune cytolytic activity, higher activation of the interferon pathway, a higher tumor mutational burden (TMB), and fewer copy number alterations (CNAs) than the other samples (all P &lt; 0.001). This new molecular subtype was named Immune Class, which served as an independent favorable prognostic factor for overall survival (hazard ratio, 0.56; 95% confidence interval, 0.33-0.97). Immune Class in cooperation with previously reported tumor and stroma classifications had a cumulative effect on PDAC prognostic stratification. Moreover, programmed cell death-1 (PD-1) inhibitors showed potential efficacy for Immune Class (P = 0.04). The robustness of our immune molecular subtypes was further verified in the validation cohort.ConclusionsBy capturing immune-related signals in the PDAC tumor microenvironment, we reveal a novel molecular subtype, Immune Class. Immune Class serves as an independent favorable prognostic factor for overall survival in PDAC patients.

Sign in / Sign up

Export Citation Format

Share Document