Diagnosis Of Lung Adenocarcinoma Research Articles

Computer-aided diagnosis of ground glass pulmonary nodule by fusing deep learning and radiomics features

Objectives. This study aims to develop a computer-aided diagnosis (CADx) scheme to classify between benign and malignant ground glass nodules (GGNs), and fuse deep leaning and radiomics imaging features to improve the classification performance. Methods. We first retrospectively collected 513 surgery histopathology confirmed GGNs from two centers. Among these GGNs, 100 were benign and 413 were malignant. All malignant tumors were stage I lung adenocarcinoma. To segment GGNs, we applied a deep convolutional neural network and residual architecture to train and build a 3D U-Net. Then, based on the pre-trained U-Net, we used a transfer learning approach to build a deep neural network (DNN) to classify between benign and malignant GGNs. With the GGN segmentation results generated by 3D U-Net, we also developed a CT radiomics model by adopting a series of image processing techniques, i.e. radiomics feature extraction, feature selection, synthetic minority over-sampling technique, and support vector machine classifier training/testing, etc. Finally, we applied an information fusion method to fuse the prediction scores generated by DNN based CADx model and CT-radiomics based model. To evaluate the proposed model performance, we conducted a comparison experiment by testing on an independent testing dataset. Results. Comparing with DNN model and radiomics model, our fusion model yielded a significant higher area under a receiver operating characteristic curve (AUC) value of 0.73 ± 0.06 (P < 0.01). The fusion model generated an accuracy of 75.6%, F1 score of 84.6%, weighted average F1 score of 70.3%, and Matthews correlation coefficient of 43.6%, which were higher than the DNN model and radiomics model individually. Conclusions. Our experimental results demonstrated that (1) applying a CADx scheme was feasible to diagnosis of early-stage lung adenocarcinoma, (2) deep image features and radiomics features provided complementary information in classifying benign and malignant GGNs, and (3) it was an effective way to build DNN model with limited dataset by using transfer learning. Thus, to build a robust image analysis based CADx model, one can combine different types of image features to decode the imaging phenotypes of GGN.

FP02.09 Effects of Chinese Reimbursement Policy on Molecular Diagnosis and Therapy of NSCLC from the Northern Part of Henan Province

Gefitinib (Irisa) have significantly prolonged the overall survival of advanced non-small cell lung cancer (NSCLC) patients with the EGFR sensitivity mutations positive (EGFR+). Our hospital, the first affiliated hospital of Xinxiang Medical University, is the largest provincial Grade 3A hospital in northern part of Henan Province, most of the patients come from rural areas, and the incomes of the patients and their family members are too low to afford the expensive antitumor costs. Therefore, the expensive costs prevent the timely and sufficient treatment with gefitinib in patients with positive EGFR+ in our hospital. Fortunately, on September 2016, gefitinib was included in health care policy of Henan Province, thus the expenses were reduced significantly. The purpose of this study was to investigate the influence of China's reimbursement policy on molecular diagnosis and treatment of advanced lung adenocarcinoma (LUAD) in our hospital. Here, the data of the LUAD from February 2015 to August 2016, as well as from September 2016 to March 2018 of our hospital were collected. The chi-square test is used to analysis: (1)the detection rate of EGFR gene mutations (EGFR-M) before and after reimbursement. (2)the detection rate of EGFR-M in the department of oncology before and after reimbursement respectively. (3) the detection rate of EGFR-M in the department of non-oncology before and after reimbursement respectively. (4)the application of EGFR-TKIs(including gefitinib and icotinib) in patients with EGFR+ before & after reimbursement. In this study, 315 cases and 405 cases were collected befor and after Semptember 2016 respectively: (1)The detection rate of EGFR-M was increased from 13.65% to 44.55% after reimbursement (P≤0.01) (Fig1A). (2)In the department of oncology, the detection rate of EGFR-M was elevated from 20.53% to 58.94% after reimbursement (P≤0.01). (3) In the department of non-oncology, the detection rate was also increased from 3.20% to 35% after reimbursement (P≤0.01). (4) The detection rate of EGFR-M was higher in the department of oncology than that of non-oncology department both before and after reimbursement (P≤0.01) (Fig1B & C). (5) After reimbursement, the application of EGFR-TKIs on patients with EGFR+ was increased prominently. There are only 21.74%(5/23) patients received treatment of EGFR-TKIs before September 2016, while most of patients with positive EGFR-sensitive mutations (72.55%, 74/102) were treated by EGFR-TKIs after September 2016(Fig1D) (P≤0.01). Chinese reimbursement policy can affect the molecular analysis and therapy of NSCLC from the northern part of Henan province significantly.

Feasibility of intraoperative diagnosis of lung adenocarcinoma in situ to avoid excessive resection.

BackgroundLimited lung resection is generally believed to be available for lung adenocarcinoma in situ (AIS). At our institute, intraoperative hematoxylin-eosin staining of frozen-section slides is routinely performed for evaluating tumor invasiveness after partial resection to avoid excessive lung resection. This study aimed to evaluate the feasibility and usefulness of intraoperative frozen-section diagnosis of AIS.MethodsWe retrospectively reviewed 143 patients with 151 AISs diagnosed by intraoperative frozen sections between 2012 and 2019 at our institute. All patients underwent limited resection because of the result of intraoperative frozen-section diagnosis.ResultsThe total concordance rate between the diagnoses of AIS by intraoperative frozen sections and postoperative paraffin-embedded sections was 82.7% for 151 nodules. Although 21 minimally invasive adenocarcinomas (MIA) and 5 invasive adenocarcinomas were diagnosed as AIS intraoperatively, no patient had tumor recurrence after resection. Among 125 pathologically proven cases of AIS postoperatively, there were 67 (53.6%) radiologically invasive tumors including ground-glass nodules (GGNs) with part-solid component or pure-solid nodules.ConclusionsThis intraoperative evaluation of frozen-section slides will help surgeons avoid excessive lung resection for AIS that was radiologically diagnosed as invasive adenocarcinoma. Intraoperative frozen-section diagnosis will provide to be clinically useful and lead to less invasive surgical treatment for lung nodules.

Comparison of real-life data from patients with NGS panel negative and KRAS mutation positive metastatic lung adenocarcinoma

To evaluate clinical and demographic characteristics and the results of cytotoxic treatments of KRASG12C, KRASother, and next-generation sequencing (NGS) panel negative patients. NGS data of 1264 patients with non-small cell lung cancer were retrospectively evaluated. Among these patients, the mutation distributions of 1081 patients with metastatic lung adenocarcinoma were analyzed. A total of 150 patients with negative NGS panel or mutant KRAS followed up in our clinic were included. Clinical features, overall survival, first-line chemotherapy responses, and progression-free survival of NGS panel negative, KRASG12C, and KRASother groups were compared. In 1081 patients who underwent NGS from tumor tissue with the diagnosis of metastatic lung adenocarcinoma, 296 (27%) NGS panel negative and 276 (26%) KRAS mutant patients were detected. Among these patients, 150 patients whose data were available were 71 (47.3%) NGS panel negative, 54 (36%) KRASother, and 25 (16.7%) KRASG12C. Clinical features, brain metastasis, and first-line chemotherapy response were similar among groups. Bone metastases were detected more often in the NGS panel negative group (p = 0.03). The median follow-up was 8.4 months. Overall, 107 deaths had occurred at the time of analysis. There was no difference in overall survival (p = 0.56) or progression-free survival (p = 0.71) among NGS panel negative, KRASother, and KRASG12C patients. There is no difference in overall survival, first-line chemotherapy response, or progression-free survival among patients with NGS panel negative, KRASG12C, or KRASother metastatic lung adenocarcinoma. Bone metastases were observed more frequently in the NGS panel negative group.

Transesophageal endoscopic ultrasound fine needle aspiration of vertebral body osteolytic tumors - a novel diagnostic approach. Case series.

Vertebral lesions, either primary or more frequently metastasis, are difficult targets for percutaneous guided biopsies and surgical biopsies and are associated with greater risks of complications. We investigated the feasibility of endoscopic ultrasound (EUS) fine needle aspiration (FNA) biopsy in the assessment of vertebral osteolytic tumors as an alternative to CT guided biopsy which is the technique currently used. Four patients with osteolytic tumors of the vertebral bodies identified by imaging methods (CT or MRI) - 3 patients, and one with a tumor detected primarily during EUS procedure were included in order to evaluate the feasibility of the procedure. The lesions were located either at the dorsal or lumbar vertebrae. In all cases we performed EUS FNA of the osteolytic vertebral body lesions with 22G needles using the transesophageal or transgastric approach. In all cases EUS FNA provided enough tissue for an accurate histopathological report, with no procedural complication. We diagnosed lung adenocarcinoma, hepatocarcinoma and a pancreatic adenocarcinoma vertebral metastasis and one case of lymphoma. EUS FNA is a valuable technique which should be considered in selected cases, when a "traditional approach" is not applicable or associated with a higher risk. Treatment guidelines are based on the histology of the tumor, histopathological examination being nowadays mandatory. Therefore, we propose for selected cases a feasible technique, with significantly lower procedural risks, as an alternative for open surgical biopsies or computed tomography guided biopsies.

The PITX gene family as potential biomarkers and therapeutic targets in lung adenocarcinoma.

The PITX gene family of transcription factors have been reported to regulate the development of multiple organs. This study was designed to investigate the role of PITXs in lung adenocarcinoma (LUAD).In this study, the transcriptional levels of the 3 identified PITXs in patients with LUAD were examined using the gene expression profiling interactive analysis interactive web server. Meanwhile, the immunohistochemical data of the 3 PITXs were obtained in the Human Protein Atlas website, and western blotting was additionally conducted for further verification. Moreover, the association between the levels of PITXs and the stage plot as well as overall survival of patients with LUAD was analyzed.We found that the mRNA and protein levels of PITX1 and PITX2 were higher in LUAD tissues than those in normal lung tissues, while those of PITX3 displayed no significant differences. Additionally, PITX1 and PITX3 were found to be significantly associated with the stage of LUAD. The Kaplan-Meier Plot showed that the high level of PITX1 conferred a better overall survival of patients with LUAD while the high level of PITX3 was associated with poor prognosis.Our study implied that PITX1 and PITX3 are potential targets of precision therapy for patients with LUAD while PITX1 and PITX2 are regarded as novel biomarkers for the diagnosis of LUAD.

Integrating NGS-derived mutational profiling in the diagnosis of multiple lung adenocarcinomas

MicroAbstractIntegration of Next Generation Sequencing (NGS) information for use in distinguishing between Multiple Primary Lung Cancer and intrapulmonary metastasis was evaluated. We used a probabilistic model, comprehensive histologic assessment and NGS to classify patients. Integrating NGS data confirmed initial diagnosis (n = 41), revised the diagnosis (n = 12), while resulted in non-informative data (n = 8). Accuracy of diagnosis can be significantly improved with integration of NGS data. BackgroundDistinguishing between multiple primary lung cancers (MPLC) and intrapulmonary metastases (IPM) is challenging. The goal of this study was to evaluate how Next Generation Sequencing (NGS) information may be integrated in the diagnostic strategy. Patients and MethodsPatients with multiple lung adenocarcinomas were classified using both the comprehensive histologic assessment and NGS. We computed the joint probability of each pair having independent mutations by chance (thus being classified as MPLC). These probabilities were computed using the marginal mutation rates of each mutation, and the known negative dependencies between driver genes and different gene loci. With these NGS-driven data, cases were re-classified as MPLC or IPM. ResultsWe analyzed 61 patients with a total of 131 tumors. The most frequent mutation was KRAS (57.3%) which occured at a rate higher than expected (p < 0.001) in lung cancer. No mutation was detected in 25/131 tumors (19.1%). Discordant molecular findings between tumor sites were found in 46 patients (75.4%); 11 patients (18.0%) had concordant molecular findings, and 4 patients (6.6%) had concordant molecular findings at 2 of the 3 sites. After integration of the NGS data, the initial diagnosis was confirmed for 41 patients (67.2%), the diagnosis was revised for 12 patients (19.7%) or was considered as non-informative for 8 patients (13.1%). ConclusionIntegrating the information of NGS data may significantly improve accuracy of diagnosis and staging.

Immunocytochemical Detection of ALK and ROS1 Rearrangements in Lung Cancer Cytological Samples.

The detection of molecular alterations such as ROS1 and ALK rearrangements is performed as part of the diagnosis of advanced-stage lung adenocarcinoma. These alterations allow the treatments with tyrosine kinase inhibitors. Cytological samples are very useful as up to 40% patients are diagnosed with this type of sample. Here we describe the immunocytochemistry technique usable to reveal the overexpression of ALK or ROS1 tyrosine kinase receptors secondary to ALK and ROS1 rearrangements, respectively.

Aberrant thyroid transcription factor-1 expression in ovarian and nasopharyngeal carcinoma: Case reports with review of literature

Introduction: Thyroid transcription factor-1 (TTF-1) is the recommended and sensitive immunohistochemical (IHC) marker for diagnosing lung adenocarcinomas and thyroid neoplasms, at the primary as well as the metastatic site, and is useful to rule out these sites in cases with adenocarcinomas of unknown primary. However, aberrant nuclear TTF-1 expression has also been reported in malignancies arising from gastrointestinal tract (GIT), breast carcinoma, female genital tract (FGT), and colon. Here, we present our experience of aberrant expression of TTF-1 in ovarian and nasopharyngeal carcinoma (NPC). Materials and Methods: TTF-1 immunostain was performed on 16 cases of NPC using clones SP141 and 8G7G3/1. A case of ovarian serous carcinoma with aberrant TTF-1 expression has also been discussed. Results: Of 16 cases of NPC, 2 cases (12.5%) showed strong TTF-1 expression with clone SP141, while all were negative with 8G7G3/1. In one of the case initial diagnosis of metastatic lung adenocarcinoma was made on cervical node biopsy as the tumor was negative for p40 and showed strong nuclear positivity for TTF-1 (SP141 clone). However, on clinical suspicion of NPC Epstein-Barr virus-encoded small RNA (EBER) by in situ hybridization was performed which was positive, and repeat TTF-1 using clone 8G7G3/1 was negative, suggested a NPC. In the case of ovarian tumor, cells were strongly positive for CK7 and TTF-1 (SP141 clone) expression, and thus the tumor was erroneously diagnosed as metastatic adenocarcinoma from a primary lung. However, IHC in the resected specimen showed strong expression for PAX-8 and TTF-1 using SP141 clone, while negative with clone 8G7G3/1 clone (Ventana), leading to final diagnosis of high-grade serous carcinoma of ovary with aberrant TTF-1 expression. In this case, addition of PAX-8 in initial biopsy diagnosis would have saved from labeling it as primary from lung. Conclusion: The sensitivity and aberrant expression vary between the various antibody clones used, with increased expression seen by SP141 clone as compared to 8G7G3/1 in our cases. Our findings highlight the significance of using a panel of diagnostic IHC markers before final diagnosis and importance of clinical details.

Neutral Desorption Extractive Electrospray Ionization Mass Spectrometry Analysis Sputum for Non-Invasive Lung Adenocarcinoma Detection.

PurposeIncreased use of low-dose spiral computed tomography (LDCT: low-dose computed tomography) screening has contributed to more frequent incidental detection of peripheral lung nodules, part of them were adenocarcinoma, which need to be further evaluated to establish a definitive diagnosis. Here, our primary objective was to evaluate the ambient mass spectrometry (AMS) sputum analysis as a non-invasive lung adenocarcinoma (LAC) diagnosis solution.Patients and MethodsNeutral desorption extractive electrospray ionization mass spectrometry (ND-EESI-MS) and collision induced dissociation (CID) were used to detect sputum metabolites from 143 spontaneous sputum samples. Partial least squares-discriminant analysis (PLS-DA) was used to refine the biomarker panel, whereas orthogonal PLS-DA (OPLS-DA) was used to operationalize the enhanced biomarker panel for diagnosis.ResultsIn this approach, 19 altered metabolites were detected by ND-EESI-MS from 76 cases of LAC and 67 cases of control. Significance testing and receiver operating characteristic (ROC) analysis identified 5 metabolites [hydroxyphenyllactic acid, phytosphingosine, N-nonanoylglycine, sphinganine, S-carboxymethyl-L-cysteine] with p <0.05 and AUC >0.75, respectively. Evaluation of model performance for prediction of LAC resulted in a cross-validation classification accuracy of 87.9%. Metabolic pathway analysis showed that sphingolipid metabolism, fatty acid metabolism, carnitine synthesis and Warburg effect were most impacted in response to disease.ConclusionThis study indicates that the application of ND-EESI-MS to sputum analysis can be used as a non-invasive detection of peripheral lung nodules. The use of sputum metabolite biomarkers may aid in the development of a further evaluation program for lung adenocarcinoma.

Analysis of the role of METTL5 as a hub gene in lung adenocarcinoma based on a weighted gene co-expression network.

Lung adenocarcinoma (LUAD) is a frequently diagnosed malignant tumor that is highly invasive and lethal. The prognosis of patients with LUAD still needs to be improved, as conventional treatment is remarkably well tolerated. In this study, the expression profile of LUAD in the TCGA database was used for differential expression analysis, and differential expression genes were determined to construct a weighted gene co-expression network analysis (WGCNA) for dividing and finding the gene modules with the highest correlation with tumor stage. Here, METTL5, DDX23, GPSM2, CEP95, WDCP, and METL17 were identified as hub genes. According to the relation degree, METTL5 was determined as the candidate gene in this study. Difference analysis and receiver operating characteristic (ROC) curve were applied to identify the predictive performance of METTL5 in LUAD, and Kaplan-Meier (KM) analysis showed that the prognosis of LUAD patients with high METTL5 expression was poor. Further GSEA analysis showed that high-expressed METTL5 was related to epithelial-mesenchymal transition and other pathways. Therefore, METTL5 may be involved in the occurrence and malignant progression of LUAD. The current findings provide an effective molecular target for early diagnosis of LUAD, helping monitor the malignant progression of LUAD and improve the prognosis of LUAD patients.

Targeted Next-Generation Sequencing Reveals Exceptionally High Rates of Molecular Driver Mutations in Never-Smokers with Lung Adenocarcinoma

Background: Historically, high rates of actionable driver mutations have been reported in never-smokers with lung adenocarcinoma (LUAD); however, in the era of modern, comprehensive cancer mutation sequencing, this relationship necessitates a more detailed analysis. Methods: All Mount Sinai patients between 01/01/2015 and 6/01/2020 with a diagnosis of LUAD of any stage with known smoking status who received genomic testing were included. Most patients were analyzed using the Sema4 hotspot panel or the OCAv3 NGS panel conducted at Sema4. Patients were considered fully genotyped if they were comprehensively analyzed for alterations in EGFR, KRAS, MET, ALK, RET, ROS1, BRAF, NTRK1-3, and ERBB2, otherwise they were considered partially genotyped.Findings236 never-smokers and 671 smokers met the above criteria. Of the never-smokers, 201 (85%) had a driver mutation with 175 (74%) considered actionable. Among smokers, 436 (65%) had an identified driver mutation with 268 (40%) actionable (p<0.0001). When comprehensively sequenced, 95% (70/74) of never-smokers had a driver mutation with 80% (59/74) actionable; whereas, for smokers, 73% (131/180) had a driver with only 44% (79/180) actionable (p<0.0001). Within mutations groups, EGFR G719X and KRAS G12Cs were more common to smokers. For stage IV patients harboring EGFR-mutant tumors treated with EGFR-directed therapies, never-smokers had significantly improved overall survival compared to smokers (HR=3.50; p=0.0066). In multivariable analysis, Asian ancestry and female gender remained significant predictors of 1) overall survival i n stage IV patients and 2) likelihood of harboring an actionable driver mutation. Interpretation: Comprehensive NGS revealed driver alterations in 95% of never-smokers, with the majority having an associated therapy available. All efforts should be exhausted to identify or rule out the presence of an actionable driver mutation in LUAD patients requiring systemic treatment. Funding: Center for Thoracic Oncology Seed funding Declaration of Interests: MIK, KLA, XZ, SG, MF, MR, HA, FMH, WKO, SN, EES, RC are employed by Sema4 and some declare ownership of stock options of Sema4. No other competing interests to declare. PCM has received honoraria from Guardant Health and Amgen. FRH served on advisory boards for Bristol-Myers Squibb, Merck, Novartis, Genentech, AstraZeneca/Daiichi, Sanofi/Regeneron, OncoCyte, Amgen. Patent: “EGFR copy number and Protein Expression as Predictive Biomarker for EGFR directed Therapy” Co-inventor. Patent through University of Colorado. (no royalties) Ethics Approval Statement: Study permission was granted by the Mount Sinai IRB. The procedures followed in this study were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration.

Radiological imaging and pathological findings of small lung adenocarcinoma: a narrative review.

The eighth edition of the Lung Cancer Handling Regulations defines the pathological findings of “invasion” in the pathological diagnosis of lung adenocarcinoma and terms it as adenocarcinoma in situ/minimally invasive carcinoma. In addition, the invasion diameter (tumor diameter excluding the lepidic growth region) was adopted as the pT factor, and the classification further reflected prognosis (degree of invasion/progression). Meanwhile, computed tomography imaging-based classification, where the consolidation (nodule) diameter excluding the ground glass shadow area was defined as cT, and the classification reflected the pathological invasion diameter. It is clear that the revision of the eighth edition has reduced discrepancies in the pathological findings of lung adenocarcinoma in CT imaging and assessment of the degree of invasion and progression. At the same time, the 8th edition is not yet accurate enough. Therefore, we will discuss imaging techniques to better predict the extent of adenocarcinoma invasion and progression, based on our own findings and the literature.

Immunoexpression of TTF1 and p63 Differentiates Lung Adenocarcinomas in Sputum Samples.

Context:Differentiating NSCLC as either adeno or squamous type and identification of Epidermal Growth Factor Receptor (EGFR) mutations is clinically relevant for lung cancer patients for selecting treatment. Thyroid transcription factor-1 (TTF-1) and p63 were demonstrated as useful markers for histologic typing of lung cancer. Mutation and overexpression of EGFR has been reported in a subset of non-small cell lung cancers. If these markers can be validated for the differential diagnosis of adenocarcinoma in a sputum sample itself, it will be highly beneficial for lung cancer patients.Aims:To evaluate whether immunocytochemical expression of TTF-1, p63, and EGFR proteins in sputum samples can be used for differential diagnosis of lung adenocarcinoma by comparing with that of the corresponding tissue samples.Settings and Design:Ninety sputum samples and matched tissue samples were used for the study.Subjects and Methods:Monolayered smears and cell blocks of sputum and the corresponding tissue samples were immunostained with the standard ABC method. The expression patterns of these markers were analyzed statistically and compared with clinic-pathological parameters.Statistical Analysis Used:Chi-square test and paired t-test.Results:The p63 protein had a positive expression in 73.9% of SCC whereas TTF1 had positive expression in 75.8% of ADC. The EGFR expression was positive in 27 cases of adenocarcinoma, 21 cases of SCC and 19 cases of NSCLC.Conclusions:Immunocytochemistry of the aforementioned antibodies in sputum samples can be used as supplementary evidence for the subtyping of NSCLC.

Identification and Validation of Two Lung Adenocarcinoma-Development Characteristic Gene Sets for Diagnosing Lung Adenocarcinoma and Predicting Prognosis.

Background: Lung adenocarcinoma (LUAD) is one of the main types of lung cancer. Because of its low early diagnosis rate, poor late prognosis, and high mortality, it is of great significance to find biomarkers for diagnosis and prognosis.Methods: Five hundred and twelve LUADs from The Cancer Genome Atlas were used for differential expression analysis and short time-series expression miner (STEM) analysis to identify the LUAD-development characteristic genes. Survival analysis was used to identify the LUAD-unfavorable genes and LUAD-favorable genes. Gene set variation analysis (GSVA) was used to score individual samples against the two gene sets. Receiver operating characteristic (ROC) curve analysis and univariate and multivariate Cox regression analysis were used to explore the diagnostic and prognostic ability of the two GSVA score systems. Two independent data sets from Gene Expression Omnibus (GEO) were used for verifying the results. Functional enrichment analysis was used to explore the potential biological functions of LUAD-unfavorable genes.Results: With the development of LUAD, 185 differentially expressed genes (DEGs) were gradually upregulated, of which 84 genes were associated with LUAD survival and named as LUAD-unfavorable gene set. While 237 DEGs were gradually downregulated, of which 39 genes were associated with LUAD survival and named as LUAD-favorable gene set. ROC curve analysis and univariate/multivariate Cox proportional hazards analyses indicated both of LUAD-unfavorable GSVA score and LUAD-favorable GSVA score were a biomarker of LUAD. Moreover, both of these two GSVA score systems were an independent factor for LUAD prognosis. The LUAD-unfavorable genes were significantly involved in p53 signaling pathway, Oocyte meiosis, and Cell cycle.Conclusion: We identified and validated two LUAD-development characteristic gene sets that not only have diagnostic value but also prognostic value. It may provide new insight for further research on LUAD.

Establishment of a Gene Signature to Predict Prognosis for Patients with Lung Adenocarcinoma.

Accumulating evidence indicates that the reliable gene signature may serve as an independent prognosis factor for lung adenocarcinoma (LUAD) diagnosis. Here, we sought to identify a risk score signature for survival prediction of LUAD patients. In the Gene Expression Omnibus (GEO) database, GSE18842, GSE75037, GSE101929, and GSE19188 mRNA expression profiles were downloaded to screen differentially expressed genes (DEGs), which were used to establish a protein-protein interaction network and perform clustering module analysis. Univariate and multivariate proportional hazards regression analyses were applied to develop and validate the gene signature based on the TCGA dataset. The signature genes were then verified on GEPIA, Oncomine, and HPA platforms. Expression levels of corresponding genes were also measured by qRT-PCR and Western blotting in HBE, A549, and PC-9 cell lines. The prognostic signature based on eight genes (TTK, HMMR, ASPM, CDCA8, KIF2C, CCNA2, CCNB2, and MKI67) was established, which was independent of other clinical factors. The risk model offered better discrimination between risk groups, and patients with high-risk scores tended to have poor survival rate at 1-, 3- and 5-year follow-up. The model also presented better survival prediction in cancer-specific cohorts of age, gender, clinical stage III/IV, primary tumor 1/2, and lymph node metastasis 1/2. The signature genes, moreover, were highly expressed in A549 and PC-9 cells. In conclusion, the risk score signature could be used for prognostic estimation and as an independent risk factor for survival prediction in patients with LUAD.

Serum microRNA expression profiling revealing potential diagnostic biomarkers for lung adenocarcinoma.

BackgroundRecent studies have demonstrated that microRNAs (miRNAs) in the blood circulation can serve as promising diagnostic markers for cancers. This four-stage study aimed at finding serum miRNAs as potential biomarkers for lung adenocarcinoma (LA) diagnosis.MethodsThe study was carried out between 2016 and 2017. The Exiqon miRNA qPCR panel (3 LA vs. 1 normal control [NC] pooled serum samples) was used for initial screening to acquire miRNA profiles. Thirty-five dysregulated miRNAs were further evaluated in the training (24 LA vs. 24 NCs) and testing stages (110 LA vs. 110 NCs) using quantitative real-time polymerase chain reaction assays.ResultsFour serum miRNAs (miR-133a-3p, miR-584-5p, miR-10b-5p, and miR-221-3p) were significantly overexpressed in LA patients compared with NCs. The diagnostic value of the four-miRNA panel was validated by an external cohort (36 LA vs. 36 NCs). The areas under the receiver operating characteristic curve of the four-miRNA panel in the training, testing, and external validation stages were 0.734, 0.803, and 0.894 respectively. Meanwhile, the expression level of miR-221-3p was much higher in LA tumor samples than that in the adjacent normal tissues (19 LA vs. 19 NCs). The expression level of miR-10b-5p was also elevated in the serum-derived exosomes samples (18 LA vs. 18 NCs). The expression of miR-133a-3p, miR-584-5p, and miR-10b-5p was significantly elevated in LA patients with epidermal growth factor receptor mutation compared with NCs.ConclusionThe study established a four-miRNA signature in serum that could improve the diagnostic capability of LA.

Deep Data Source Fusion with Bias-Undoing for Lung Adenocarcinoma Classification

Computer-aided diagnosis of early-stage lung adenocarcinoma based on deep learning is prospective for assisting the prevention and treatment of the deathly disease of lung cancer, however, relevant works face the problem of limited training data. The technique of data source fusion with the training of deep models on multiple relevant datasets is promising to resolve the lack of training data, while the bias of data distribution from different data sources exists as a universal issue to affect the learning performance. In this paper, we propose a deep learning framework based on bias-undoing data source fusion to classify early stages of lung adenocarcinoma in computed tomography (CT) images. The framework conducts learning on the integrated datasets for respectively natural image, lung nodule CT and lung adenocarcinoma CT, as designed with an organization of base parameters and bias parameters to adapt to the data distribution with bias. Experimental results demonstrate that the proposed bias-undoing framework is effective to improve the performance of deep learning for lung adenocarcinoma classification, and is with great superiority to those general fusion frameworks on alleviating the effect of dataset bias.

Role of Napsin A and Survivin Immunohistochemical Expression in Bronchogenic Adenocarcinoma.

Background:Lung cancer, being the leading cause of cancer deaths with most patients diagnosed at a late stage, represents a major burden in developing countries especially with both air pollution and tobacco use increasing. With the evolution of new, successful therapies that target lung adenocarcinoma, it became of utmost importance to diagnose lung adenocarcinoma. Despite considering TTF-1 as the predominant marker for identifying lung adenocarcinoma but it has limited sensitivity and specificity, which means that its expression decreases in relation to the degree of tumor differentiation. Aim of work:this study intended to evaluate the use of Napsin A in lung adenocarcinoma, and observe if it can withstand along the different lines of tumor differentiation and Survivin as a marker of poor prognosis. Materials and Methods:Forty paraffin blocks of bronchogenic carcinoma were collected and studied immunohistochemically against Napsin A and Survivin. Results:There was a statistically significant relation between Napsin A reactivity and tumor grade as 72% of grade II as well as all cases of grade I were strongly positive compared to none of grade III cases. Another statistically significant relation between Survivin reactivity and tumor grade was observed as all grades I and II cases showed labeling index <10%, while all grade III cases showed labeling index >10%. Conclusion: Napsin A is a good prognostic marker while Survivin stands as a poor one for lung adenocarcinoma with a statistically inverse relation between the two, which means that Napsin A can’t be used as a marker for diagnosing poorly differentiated tumors.

Evaluation of the screening and diagnosis model for the lung adenocarcinoma protein function module based on WGCNA and machine learning

This paper aims to identify lung adenocarcinoma biomarkers through WGCNA and machine learning and construct a clinical diagnosis model for lung adenocarcinoma. We used the lung cancer protein expression data from the CPTAC database to conduct differentiation analysis, built a WGCNA network of lung adenocarcinoma samples and a WGCNA network of lung tumor samples and normal samples, and assessed the overlapped module of these two networks using machine learning. GO and KEGG abundance analysis was conducted to find proteins related to lung tumor, a correlation network for proteins in the overlapped module was created to mine the target protein biomarkers, and machine learning was used to create and analyze a screening model. Therewere2317 differentially expressed proteins obtained from 213 lung tumor samples from the CPTAC database; the lung tumor network and the lung tumor para-carcinoma tissue joint network had two overlapping modules; through PPI network optimization, we mined 11 protein biomarkers related to lung adenocarcinoma. Through verification based on data outside TCGA, we found that the lung adenocarcinoma diagnosis model constructed based on the 11 protein biomarkers had high accuracy and stability, showing clinical and biological significance.