scholarly journals Classification and Mutation Prediction from Non-Small Cell Lung Cancer Histopathology Images using Deep Learning

2017 ◽  
Author(s):  
Nicolas Coudray ◽  
Andre L. Moreira ◽  
Theodore Sakellaropoulos ◽  
David Fenyö ◽  
Narges Razavian ◽  
...  
Keyword(s):  
Neural Network ◽  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Deep Learning ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Visual Analysis ◽  
The Cancer Genome Atlas ◽  
Normal Lung ◽  
Histopathology Images

AbstractVisual analysis of histopathology slides of lung cell tissues is one of the main methods used by pathologists to assess the stage, types and sub-types of lung cancers. Adenocarcinoma and squamous cell carcinoma are two most prevalent sub-types of lung cancer, but their distinction can be challenging and time-consuming even for the expert eye. In this study, we trained a deep learning convolutional neural network (CNN) model (inception v3) on histopathology images obtained from The Cancer Genome Atlas (TCGA) to accurately classify whole-slide pathology images into adenocarcinoma, squamous cell carcinoma or normal lung tissue. Our method slightly outperforms a human pathologist, achieving better sensitivity and specificity, with ∼0.97 average Area Under the Curve (AUC) on a held-out population of whole-slide scans. Furthermore, we trained the neural network to predict the ten most commonly mutated genes in lung adenocarcinoma. We found that six of these genes – STK11, EGFR, FAT1, SETBP1, KRAS and TP53 – can be predicted from pathology images with an accuracy ranging from 0.733 to 0.856, as measured by the AUC on the held-out population. These findings suggest that deep learning models can offer both specialists and patients a fast, accurate and inexpensive detection of cancer types or gene mutations, and thus have a significant impact on cancer treatment.

scholarly journals UBE2D1 RNA Expression Was an Independent Unfavorable Prognostic Indicator in Lung Adenocarcinoma, but Not in Lung Squamous Cell Carcinoma

2018 ◽  
Vol 2018 ◽  
pp. 1-8 ◽  
Author(s):  
Liyan Hou ◽  
Yingbo Li ◽  
Ying Wang ◽  
Dongqiang Xu ◽  
Hailing Cui ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Lung Adenocarcinoma ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Survival Data ◽  
Lung Squamous Cell Carcinoma ◽  
Prognostic Indicator ◽  
The Cancer Genome Atlas ◽  
Rna Expression

In this study, we investigated the potential prognostic value of ubiquitin-conjugating enzyme E2D1 (UBE2D1) RNA expression in different histological subtypes of non-small-cell lung cancer (NSCLC). A retrospective study was performed by using molecular, clinicopathological, and survival data in the Cancer Genome Atlas (TCGA)—Lung Cancer. Results showed that both lung adenocarcinoma (LUAD) (N=514) and lung squamous cell carcinoma (LUSC) (N=502) tissues had significantly elevated UBE2D1 RNA expression compared to the normal tissues (p<0.001 and p=0.036, respectively). UBE2D1 RNA expression was significantly higher in LUAD than in LUSC tissues. Increased UBE2D1 RNA expression was independently associated with shorter OS (HR: 1.359, 95% CI: 1.031–1.791, p=0.029) and RFS (HR: 1.842, 95% CI: 1.353–2.508, p<0.001) in LUAD patients, but not in LUSC patients. DNA amplification was common in LUAD patients (88/551, 16.0%) and was associated with significantly upregulated UBE2D1 RNA expression. Based on these findings, we infer that UBE2D1 RNA expression might only serve as an independent prognostic indicator of unfavorable OS and RFS in LUAD, but not in LUSC.

Machine learning analysis of DNA methylation profiles distinguishes primary lung squamous cell carcinomas from head and neck metastases

2019 ◽  
Vol 11 (509) ◽  
pp. eaaw8513 ◽  
Author(s):  
Philipp Jurmeister ◽  
Michael Bockmayr ◽  
Philipp Seegerer ◽  
Teresa Bockmayr ◽  
Denise Treue ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Dna Methylation ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Squamous Cell ◽  
Pulmonary Metastases ◽  
Normal Lung ◽  
Support Vector

Head and neck squamous cell carcinoma (HNSC) patients are at risk of suffering from both pulmonary metastases or a second squamous cell carcinoma of the lung (LUSC). Differentiating pulmonary metastases from primary lung cancers is of high clinical importance, but not possible in most cases with current diagnostics. To address this, we performed DNA methylation profiling of primary tumors and trained three different machine learning methods to distinguish metastatic HNSC from primary LUSC. We developed an artificial neural network that correctly classified 96.4% of the cases in a validation cohort of 279 patients with HNSC and LUSC as well as normal lung controls, outperforming support vector machines (95.7%) and random forests (87.8%). Prediction accuracies of more than 99% were achieved for 92.1% (neural network), 90% (support vector machine), and 43% (random forest) of these cases by applying thresholds to the resulting probability scores and excluding samples with low confidence. As independent clinical validation of the approach, we analyzed a series of 51 patients with a history of HNSC and a second lung tumor, demonstrating the correct classifications based on clinicopathological properties. In summary, our approach may facilitate the reliable diagnostic differentiation of pulmonary metastases of HNSC from primary LUSC to guide therapeutic decisions.

scholarly journals The bioinformatics analysis of RIOX2 gene in lung adenocarcinoma and squamous cell carcinoma

PLoS ONE ◽  
2021 ◽  
Vol 16 (12) ◽  
pp. e0259447
Author(s):  
Bingqing Sun ◽  
Hongwen Zhao
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Lung Adenocarcinoma ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Lung Squamous Cell Carcinoma ◽  
The Cancer Genome Atlas ◽  
Public Health Issue ◽  
High Morbidity ◽  
Link Type

Lung cancer is characterized by high morbidity and mortality rates, and it has become an important public health issue worldwide. The occurrence and development of tumors is a multi-gene and multi-stage complex process. As an oncogene, ribosomal oxygenase 2 (RIOX2) has been associated with a variety of cancers. In this article, we analyzed the correlation between RIOX2 expression and methylation in lung cancer based on the databases including the cancer genome atlas (TCGA) (https://portal.gdc.cancer.gov/) and the gene expression omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/). It was found that RIOX2 is highly expressed in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) tissues, whose expression is negatively correlated with its methylation level. In this regard, methylation at cg09716038, cg14773523, cg14941179, and cg22299097 had a significant negative correlation with RIOX2 expression in LUAD, whereas in LUSC, methylation at cg09716038, cg14773523, cg14941179, cg22299097, cg05451573, cg10779801, and cg23629183 is negatively correlated with RIOX2 expression. According to the analysis based on the databases, RIOX2 gene could not be considered as the independent prognostic biomarker in lung adenocarcinoma or squamous cell lung cancer. However, the molecular mechanism of RIOX2 gene in the development of lung cancer may be helpful in improving lung cancer therapy.

Comprehensive genomic characterization of squamous cell carcinoma of the lung.

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. 7006-7006 ◽  
Author(s):  
Ramaswamy Govindan ◽  
Peter S. Hammerman ◽  
David N. Hayes ◽  
Matthew D Wilkerson ◽  
Stephen Baylin ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Mrna Expression ◽  
Squamous Cell ◽  
Copy Number ◽  
Molecular Subtype ◽  
The Cancer Genome Atlas ◽  
Loss Of Function ◽  
Global Methylation

7006 Background: A third of patients with non-small cell lung cancer are diagnosed with squamous cell carcinoma (SCC) histology. This report describes findings from the comprehensive genomic analyses of 178 SCC samples. Methods: The Cancer Genome Atlas (TCGA) is conducting DNA, RNA, and miRNA sequencing along with DNA copy number profiling, quantification of mRNA expression and promoter methylation on surgically resected samples from previously untreated patients with stage I-III SCC of the lung. Results: The demographics of 178 patients enrolled in the study: median age 68 years (range: 40-85); female 47 (26%) and history of tobacco smoking 171 (96%). Over 30 sites of significant somatic copy number alteration (SCNA) were identified. Exome sequencing of 178 lung SCC and matched normal samples revealed 13 significantly mutated genes with a False Discovery Rate (FDR) of <0.01 and high expression levels, including TP53, CDKN2A, PTEN, KEAP1, and NFE2L2. Apart from the near universal loss of TP53 and CDKN2A, alterations in the NFE2L2/KEAP1 and PI3K/AKT pathways were found in 35% and 43% of tumors analyzed. mRNA expression profiling revealed four distinct expression subtypes, each one enriched with distinct mutations and SCNAs - classical (37%): NFE2L2 and KEAP1 mutations, FGFR kinase alterations, increased global methylation and the highest rate of tobacco use; basal (24%): alterations in FGFR kinases; secretory (24%): PDGFRA alterations; primitive (15%): RB1 mutations. Rearrangements involving several known tumor suppressors were detected by whole genome shotgun sequencing of 20 tumor/normal pairs and confirmed by RNA sequencing including PTEN, RB1, NOTCH1, NF1 and CDKN2A. CDKN2A loss by one of several mechanisms (deletion, mutation, rearrangement with loss of function and methylation) was observed in 72% of specimens. Potential therapeutic targets for clinical trials with currently available drugs were identified in 127 patients (75%). Conclusions: SCC of the lung is a distinct molecular subtype of lung cancer potentially amenable to distinct molecularly targeted therapies.

scholarly journals CLCA2 as a Novel Immunohistochemical Marker for Differential Diagnosis of Squamous Cell Carcinoma from Adenocarcinoma of the Lung

2014 ◽  
Vol 2014 ◽  
pp. 1-11 ◽  
Author(s):  
Kazuya Shinmura ◽  
Hisaki Igarashi ◽  
Hisami Kato ◽  
Yuichi Kawanishi ◽  
Yusuke Inoue ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Differential Diagnosis ◽  
Protein Expression ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Primary Lung Cancer ◽  
The Cancer Genome Atlas ◽  
Immunohistochemical Marker ◽  
Expression Status

Recent progress in targeted therapy for lung cancer has revealed that accurate differential diagnosis between squamous cell carcinoma (SCC) and adenocarcinoma (ADC) of the lung is essential. To identify a novel immunohistochemical marker useful for differential diagnosis between the two subtypes of lung cancer, we first selected 24 SCC-specific genes and 6 ADC-specific genes using data (case number, 980) from the Cancer Genome Atlas (TCGA) database. Among the genes, we chose theCLCA2gene, which is involved in chloride conductance and whose protein expression in lung cancer is yet to be characterized, and evaluated its protein expression status in 396 cases of primary lung cancer at Hamamatsu University Hospital. Immunohistochemical analysis revealed a significantly higher CLCA2 expression level in the SCCs than in the ADCs(P<0.0001)and also a significantly higher frequency of CLCA2 protein expression in the SCCs (104/161, 64.6%) as compared with that in the ADCs (2/235, 0.9%)(P<0.0001; sensitivity 64.6%, specificity 99.1%). The CLCA2 protein expression status was associated with the histological tumor grade in the SCCs. These results suggest that CLCA2 might be a novel excellent immunohistochemical marker for differentiating between primary SCC and primary ADC of the lung.

Smoking is Associated with Lung Adenocarcinoma and Lung Squamous Cell Carcinoma Progression through Inducing Distinguishing lncRNA Alterations in Different Genders

2021 ◽  
Vol 21 ◽  
Author(s):  
Bing Liu ◽  
Yuan Liu ◽  
Jingfeng Zou ◽  
Menglin Zou ◽  
Zhenshun Cheng
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Lung Adenocarcinoma ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Disease Free Survival ◽  
Lung Squamous Cell Carcinoma ◽  
The Cancer Genome Atlas ◽  
Male And Female ◽  
Smoking Index

Background: Smoking participates in pathogenesis of lung cancer. Long non-coding RNAs (lncRNAs) play some specific roles during development of lung cancers. Objective: To investigate effects of smoking on lncRNA alterations in lung cancer. Methods: There are 522 lung adenocarcinoma (LUAD) and 504 lung squamous cell carcinoma (LUSC) participants. Clinical and lncRNA genetic data were downloaded from The Cancer Genome Atlas (TCGA) database. LncRNA alterations were analyzed in lung cancer patients. Smoking category and packs were evaluated. Correlations between smoking and LncRNA alterations were analyzed. Kaplan-Meier analysis was performed to determine overall survival and disease free survival. Results: There are more non-smokers in LUSC than in LUAD. In both LUAD and LUSC, smoking could increase total mutation counts and fraction of copy number alterations. Smoking index positively correlated with total mutations in LUAD, but not in LUSC. Smoking could trigger lncRNA alterations both in LUAD and LUSC. Smoking regulated different lncRNA between male and female. EXOC3-AS1 and LINC00603 alterations were positively correlated with smoking index in male LUAD smokers. In female LUAD smokers, smoking index was positively correlated with SNHG15, TP53TG1 and LINC01600 and negatively with LINC00609 and PTCSC3. In both male and female LUSC patients, smoking increased or decreased several lncRNA alterations. DGCR5 alteration increased in male LUSC than in female LUSC patients. In female LUSC patients, LOH12CR2 alteration was positively correlated with smoking index. Conclusions: Smoking promoted LUAD and LUSC development by affecting different lncRNA alterations in different genders.

scholarly journals Diagnosis value of aberrantly expressed microRNA profiles in lung squamous cell carcinoma: a study based on the Cancer Genome Atlas

PeerJ ◽  
2017 ◽  
Vol 5 ◽  
pp. e4101 ◽  
Author(s):  
Sheng Yang ◽  
Jing Sui ◽  
Geyu Liang
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Bioinformatics Analysis ◽  
Lung Squamous Cell Carcinoma ◽  
Clinical Information ◽  
The Cancer Genome Atlas ◽  
Cancer Genome Atlas ◽  
Genome Atlas

Background Lung cancer is considered as one of the most frequent and deadly cancers with high mortality all around the world. It is critical to find new biomarkers for early diagnosis of lung cancer, especially lung squamous cell carcinoma (LUSC). The Cancer Genome Atlas (TCGA) is a database which provides both cancer and clinical information. This study is a comprehensive analysis of a novel diagnostic biomarker for LUSC, based on TCGA. Methods and Results The present study investigated LUSC-specific key microRNAs (miRNAs) from large-scale samples in TCGA. According to exclusion criteria and inclusion criteria, the expression profiles of miRNAs with related clinical information of 332 LUSC patients were obtained. Most aberrantly expressed miRNAs were identified between tumor and normal samples. Forty-two LUSC-specific intersection miRNAs (fold change >2, p < 0.05) were obtained by an integrative computational method, among them six miRNAs were found to be aberrantly expressed concerning characteristics of patients (gender, lymphatic metastasis, patient outcome assessment) through Student t-test. Five miRNAs correlated with overall survival (log-rank p < 0.05) were obtained through the univariate Cox proportional hazards regression model and Mantel–Haenszel test. Then, five miRNAs were randomly selected to validate the expression in 47 LUSC patient tissues using quantitative real-time polymerase chain reaction. The results showed that the test findings were consistent with the TCGA findings. Also, the diagnostic value of the specific key miRNAs was determined by areas under receiver operating characteristic curves. Finally, 577 interaction mRNAs as the targets of 42 LUSC-specific intersection miRNAs were selected for further bioinformatics analysis. Conclusion This study indicates that this novel microRNA expression signature may be a useful biomarker of the diagnosis for LUSC patients, based on bioinformatics analysis.

scholarly journals Interaction between the microbiome and TP53 in human lung cancer

2018 ◽  
Vol 19 (1) ◽  
Author(s):  
K. Leigh Greathouse ◽  
James R. White ◽  
Ashely J. Vargas ◽  
Valery V. Bliskovsky ◽  
Jessica A. Beck ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Cigarette Smoke ◽  
Squamous Cell ◽  
Human Lung Cancer ◽  
Normal Lung ◽  
Lung Carcinogenesis ◽  
Bacterial Gene ◽  
Tp53 Mutations

Abstract Background Lung cancer is the leading cancer diagnosis worldwide and the number one cause of cancer deaths. Exposure to cigarette smoke, the primary risk factor in lung cancer, reduces epithelial barrier integrity and increases susceptibility to infections. Herein, we hypothesize that somatic mutations together with cigarette smoke generate a dysbiotic microbiota that is associated with lung carcinogenesis. Using lung tissue from 33 controls and 143 cancer cases, we conduct 16S ribosomal RNA (rRNA) bacterial gene sequencing, with RNA-sequencing data from lung cancer cases in The Cancer Genome Atlas serving as the validation cohort. Results Overall, we demonstrate a lower alpha diversity in normal lung as compared to non-tumor adjacent or tumor tissue. In squamous cell carcinoma specifically, a separate group of taxa are identified, in which Acidovorax is enriched in smokers. Acidovorax temporans is identified within tumor sections by fluorescent in situ hybridization and confirmed by two separate 16S rRNA strategies. Further, these taxa, including Acidovorax, exhibit higher abundance among the subset of squamous cell carcinoma cases with TP53 mutations, an association not seen in adenocarcinomas. Conclusions The results of this comprehensive study show both microbiome-gene and microbiome-exposure interactions in squamous cell carcinoma lung cancer tissue. Specifically, tumors harboring TP53 mutations, which can impair epithelial function, have a unique bacterial consortium that is higher in relative abundance in smoking-associated tumors of this type. Given the significant need for clinical diagnostic tools in lung cancer, this study may provide novel biomarkers for early detection.

scholarly journals Prognostic Value of Telomeric Zinc Finger-Associated Protein Expression in Adenocarcinoma and Squamous Cell Carcinoma of Lung

Medicina ◽  
2021 ◽  
Vol 57 (11) ◽  
pp. 1223
Author(s):  
Gun-Jik Kim ◽  
Jae-Ho Lee ◽  
Mincheol Chae ◽  
Deok-Heon Lee
Keyword(s):  
Lung Cancer ◽  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Zinc Finger ◽  
Squamous Cell ◽  
Prognostic Value ◽  
Disease Free Survival ◽  
The Cancer Genome Atlas ◽  
Normal Tissues ◽  
Telomere Regulation

Background and Objectives: Telomeric zinc finger-associated protein (TZAP) is a telomere regulation protein, previously known as ZBTB48. It binds preferentially to elongated telomeres, competing with telomeric repeat factors 1 and 2. TZAP expression may be associated with carcinogenesis, however; this study has not yet been performed in lung cancer. In this study, we examined the clinicopathological and prognostic values of TZAP expression in non-small cell lung cancer (NSCLC). Materials and Methods: Data were collected from The Cancer Genome Atlas. The clinical and prognostic values of TZAP for NSCLC were examined in adenocarcinoma (AD) and squamous cell carcinoma (SCC). Results: TZAP expression significantly increased in NSCLC tissues compared with normal tissues. In AD, TZAP expression was lower in patients with higher T stage (p = 0.005), and was associated with lymph node stage in SCC (p = 0.005). Survival analysis showed shorter disease-free survival in AD patients with lower TZAP expression (p = 0.047). TZAP expression did not have other clinical or prognostic value for AD and SCC. Conclusions: TZAP expression is a potential prognostic marker for NSCLC, especially in patients with AD.

scholarly journals JOSD1 promotes proliferation and chemoresistance of head and neck squamous cell carcinoma under the epigenetic regulation of BRD4

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Chao Jing ◽  
Dandan Liu ◽  
Qingchuan Lai ◽  
Linqi Li ◽  
Mengqian Zhou ◽  
...  
Keyword(s):  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Head And Neck ◽  
Squamous Cell ◽  
Gene Expression Omnibus ◽  
Neck Squamous Cell Carcinoma ◽  
The Cancer Genome Atlas ◽  
In Vivo Experiments

Abstract Background Deubiquitinating enzymes (DUBs) play critical roles in various cancers by modulating functional proteins post-translationally. Previous studies have demonstrated that DUB Josephin Domain Containing 1 (JOSD1) is implicated in tumor progression, however, the role and mechanism of JOSD1 in head and neck squamous cell carcinoma (HNSCC) remain to be explored. In this study, we aimed to identify the clinical significance and function of JOSD1 in HNSCC. Methods The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases were analyzed to find novel DUBs in HNSCC. Immunohistochemistry assay was performed to determine the expression of JOSD1 in our cohort of 42 patients suffered with HNSCC. Kaplan–Meier analysis was used to identify the correlation between JOSD1 and the prognosis of HNSCC patients. The regulation of BRD4 on JOSD1 was determined by using pharmacological inhibition and gene depletion. The in vitro and in vivo experiments were conducted to elucidate the role of JOSD1 in HNSCC. Results The results of IHC showed that JOSD1 was aberrantly expressed in HNSCC specimens, especially in the chemoresistant ones. The overexpression of JOSD1 indicated poor clinical outcome of HNSCC patients. Moreover, JOSD1 depletion dramatically impaired cell proliferation and colony formation, and promoted cisplatin-induced apoptosis of HNSCC cells in vitro. Additionally, JOSD1 suppression inhibited the tumor growth and improved chemosensitivity in vivo. The epigenetic regulator BRD4 contributed to the upregulation of JOSD1 in HNSCC. Conclusions These results demonstrate that JOSD1 functions as an oncogene in HNSCC progression, and provide a promising target for clinical diagnosis and therapy of HNSCC.

Sign in / Sign up

Export Citation Format

Share Document