Novel Insights Into the International Association for the Study of Lung Cancer Grading System for Lung Adenocarcinoma.

The new grading system for lung adenocarcinoma proposed by the International Association for the Study of Lung Cancer (IASLC) defines prognostic subgroups on the basis of histologic patterns observed on surgical specimens. This study sought to provide novel insights into the IASLC grading system, with particular focus on recurrence-specific survival (RSS) and lung cancer-specific survival among patients with stage I adenocarcinoma. Under the IASLC grading system, tumors were classified as grade 1 (lepidic predominant with <20% high-grade patterns [micropapillary, solid, and complex glandular]), grade 2 (acinar or papillary predominant with <20% high-grade patterns), or grade 3 (≥20% high-grade patterns). Kaplan-Meier survival estimates, pathologic features, and genomic profiles were investigated for patients whose disease was reclassified into a higher grade under the IASLC grading system on the basis of the hypothesis that they would strongly resemble patients with predominant high-grade tumors. Overall, 423 (29%) of 1443 patients with grade 1 or 2 tumors classified based on the predominant pattern-based grading system had their tumors upgraded to grade 3 based on the IASLC grading system. The RSS curves for patients with upgraded tumors were significantly different from those for patients with grade 1 or 2 tumors (log-rank P < .001) but not from those for patients with predominant high-grade patterns (P = .3). Patients with upgraded tumors had a similar incidence of visceral pleural invasion and spread of tumor through air spaces as patients with predominant high-grade patterns. In multivariable models, the IASLC grading system remained significantly associated with RSS and lung cancer-specific survival after adjustment for aggressive pathologic features such as visceral pleural invasion and spread of tumor through air spaces. The IASLC grading system outperforms the predominant pattern-based grading system and appropriately reclassifies tumors into higher grades with worse prognosis, even after other pathologic features of aggressiveness are considered.

Establishment of artificial intelligence model for precise histological subtyping of lung adenocarcinoma and its application to quantitative and spatial analysis.

BACKGROUND: The histological subtype of lung adenocarcinoma is a major prognostic factor. We developed a new artificial intelligence model to classify lung adenocarcinoma images into seven histological subtypes and adopted the model for whole-slide images to investigate the relationship between the distribution of histological subtypes and clinicopathological factors. METHODS: Using histological subtype images, which are typical for pathologists, we trained and validated an artificial intelligence model. Then, the model was applied to whole-slide images of resected lung adenocarcinoma specimens from 147 cases. RESULT: The model achieved an accuracy of 99.7% in training sets and 90.4% in validation sets consisting of typical tiles of histological subtyping for pathologists. When the model was applied to whole-slide images, the predominant subtype according to the artificial intelligence model classification matched that determined by pathologists in 75.5% of cases. The predominant subtype and tumor grade (using the WHO fourth and fifth classifications) determined by the artificial intelligence model resulted in similar recurrence-free survival curves to those determined by pathologists. Furthermore, we stratified the recurrence-free survival curves for patients with different proportions of high-grade components (solid, micropapillary and cribriform) according to the physical distribution of the high-grade component. The results suggested that tumors with centrally located high-grade components had a higher malignant potential (P < 0.001 for 5-20% high-grade component). CONCLUSION: The new artificial intelligence model for histological subtyping of lung adenocarcinoma achieved high accuracy, and subtype quantification and subtype distribution analyses could be achieved. Artificial intelligence model therefore has potential for clinical application for both quantification and spatial analysis.

Elastin in pulmonary pathology: relevance in tumours with a lepidic or papillary appearance. A comprehensive understanding from a morphological viewpoint.

Elastin and collagen are the main components of the lung connective tissue network, and together provide the lung with elasticity and tensile strength. In pulmonary pathology, elastin staining is used to variable extents in different countries. These uses include evaluation of the pleura in staging, and the distinction of invasion from collapse of alveoli after surgery (iatrogenic collapse). In the latter, elastin staining is used to highlight distorted but pre-existing alveolar architecture from true invasion. In addition to variable levels of use and experience, the interpretation of elastin staining in some adenocarcinomas leads to interpretative differences between collapsed lepidic patterns and true papillary patterns. This review aims to summarise the existing data on the use of elastin staining in pulmonary pathology, on the basis of literature data and morphological characteristics. The effect of iatrogenic collapse and the interpretation of elastin staining in pulmonary adenocarcinomas is discussed in detail, especially for the distinction between lepidic patterns and papillary carcinoma.

Characteristic of molecular subtypes in lung adenocarcinoma based on m6A RNA methylation modification and immune microenvironment.

BACKGROUND: Lung adenocarcinoma (LUAD) is a major subtype of lung cancer and closely associated with poor prognosis. N6-methyladenosine (m6A), one of the most predominant modifications in mRNAs, is found to participate in tumorigenesis. However, the potential function of m6A RNA methylation in the tumor immune microenvironment is still murky. METHODS: The gene expression profile cohort and its corresponding clinical data of LUAD patients were downloaded from TCGA database and GEO database. Based on the expression of 21 m6A regulators, we identified two distinct subgroups by consensus clustering. The single-sample gene-set enrichment analysis (ssGSEA) algorithm was conducted to quantify the relative abundance of the fraction of 28 immune cell types. The prognostic model was constructed by Lasso Cox regression. Survival analysis and receiver operating characteristic (ROC) curves were used to evaluate the prognostic model. RESULT: Consensus classification separated the patients into two clusters (clusters 1 and 2). Those patients in cluster 1 showed a better prognosis and were related to higher immune scores and more immune cell infiltration. Subsequently, 457 differentially expressed genes (DEGs) between the two clusters were identified, and then a seven-gene prognostic model was constricted. The survival analysis showed poor prognosis in patients with high-risk score. The ROC curve confirmed the predictive accuracy of this prognostic risk signature. Besides, further analysis indicated that there were significant differences between the high-risk and low-risk groups in stages, status, clustering subtypes, and immunoscore. Low-risk group was related to higher immune score, more immune cell infiltration, and lower clinical stages. Moreover, multivariate analysis revealed that this prognostic model might be a powerful prognostic predictor for LUAD. Ultimately, the efficacy of this prognostic model was successfully validated in several external cohorts (GSE30219, GSE50081 and GSE72094). CONCLUSION: Our study provides a robust signature for predicting patients' prognosis, which might be helpful for therapeutic strategies discovery of LUAD.

Multi-scale supervised clustering-based feature selection for tumor classification and identification of biomarkers and targets on genomic data.

BACKGROUND: The small number of samples and the curse of dimensionality hamper the better application of deep learning techniques for disease classification. Additionally, the performance of clustering-based feature selection algorithms is still far from being satisfactory due to their limitation in using unsupervised learning methods. To enhance interpretability and overcome this problem, we developed a novel feature selection algorithm. In the meantime, complex genomic data brought great challenges for the identification of biomarkers and therapeutic targets. The current some feature selection methods have the problem of low sensitivity and specificity in this field. RESULTS: In this article, we designed a multi-scale clustering-based feature selection algorithm named MCBFS which simultaneously performs feature selection and model learning for genomic data analysis. The experimental results demonstrated that MCBFS is robust and effective by comparing it with seven benchmark and six state-of-the-art supervised methods on eight data sets. The visualization results and the statistical test showed that MCBFS can capture the informative genes and improve the interpretability and visualization of tumor gene expression and single-cell sequencing data. Additionally, we developed a general framework named McbfsNW using gene expression data and protein interaction data to identify robust biomarkers and therapeutic targets for diagnosis and therapy of diseases. The framework incorporates the MCBFS algorithm, network recognition ensemble algorithm and feature selection wrapper. McbfsNW has been applied to the lung adenocarcinoma (LUAD) data sets. The preliminary results demonstrated that higher prediction results can be attained by identified biomarkers on the independent LUAD data set, and we also structured a drug-target network which may be good for LUAD therapy. CONCLUSIONS: The proposed novel feature selection method is robust and effective for gene selection, classification, and visualization. The framework McbfsNW is practical and helpful for the identification of biomarkers and targets on genomic data. It is believed that the same methods and principles are extensible and applicable to other different kinds of data sets.

Tumor microenvironment characterization identifies two lung adenocarcinoma subtypes with specific immune and metabolic state.

The tumor microenvironment (TME) is a vital component of tumor tissue. Increasing evidence suggests their significance in predicting outcomes and guiding therapies. However, no studies have reported a systematic analysis of the clinicopathologic significance of TME in lung adenocarcinoma (LUAD). Here, we inferred tumor stromal cells in 1184 LUAD patients using computational algorithms based on bulk tumor expression data, and evaluated the clinicopathologic significance of stromal cells. We found LUAD patients showed heterogeneous abundance in stromal cells. Infiltration of stromal cells was influenced by clinicopathologic features, such as age, gender, smoking, and TNM stage. By clustering stromal cells, we identified 2 clinically and molecularly distinct LUAD subtypes with immune active and immune repressed features. The immune active subtype is characterized by repressed metabolism and repressed proliferation of tumor cells, while the immune repressed subtype is characterized by active metabolism and active proliferation of tumor cells. Differentially expressed gene analysis of the two LUAD subtypes identified an immune activation signature. To diagnose TME subtypes practically, we constructed a TME score using principal component analysis based on the immune activation signature. The TME score predicted TME subtypes effectively in 3 independent datasets with areas under the receiver operating characteristic curves of 0.960, 0.812, and 0.819, respectively. In conclusion, we proposed 2 clinically and molecularly distinct LUAD subtypes based on tumor microenvironment that could be valuable in predicting clinical outcome and guiding immunotherapy.

Reciprocal expression of trefoil factor-1 and thyroid transcription factor-1 in lung adenocarcinomas.

Molecular targeted therapies against EGFR and ALK have improved the quality of life of lung adenocarcinoma patients. However, targetable driver mutations are mainly found in thyroid transcription factor-1 (TTF-1)/NK2 homeobox 1 (NKX2-1)-positive terminal respiratory unit (TRU) types and rarely in non-TRU types. To elucidate the molecular characteristics of the major subtypes of non-TRU-type adenocarcinomas, we analyzed 19 lung adenocarcinoma cell lines (11 TRU types and 8 non-TRU types). A characteristic of non-TRU-type cell lines was the strong expression of TFF-1 (trefoil factor-1), a gastric mucosal protective factor. An immunohistochemical analysis of 238 primary lung adenocarcinomas resected at Jichi Medical University Hospital revealed that TFF-1 was positive in 31 cases (13%). Expression of TFF-1 was frequently detected in invasive mucinous (14/15, 93%), enteric (2/2, 100%), and colloid (1/1, 100%) adenocarcinomas, less frequent in acinar (5/24, 21%), papillary (7/120, 6%), and solid (2/43, 5%) adenocarcinomas, and negative in micropapillary (0/1, 0%), lepidic (0/23, 0%), and microinvasive adenocarcinomas or adenocarcinoma in situ (0/9, 0%). Expression of TFF-1 correlated with the expression of HNF4-α and MUC5AC (P < .0001, P < .0001, respectively) and inversely correlated with that of TTF-1/NKX2-1 (P < .0001). These results indicate that TFF-1 is characteristically expressed in non-TRU-type adenocarcinomas with gastrointestinal features. The TFF-1-positive cases harbored KRAS mutations at a high frequency, but no EGFR or ALK mutations. Expression of TFF-1 correlated with tumor spread through air spaces, and a poor prognosis in advanced stages. Moreover, the knockdown of TFF-1 inhibited cell proliferation and soft-agar colony formation and induced apoptosis in a TFF-1-high and KRAS-mutated lung adenocarcinoma cell line. These results indicate that TFF-1 is not only a biomarker, but also a potential molecular target for non-TRU-type lung adenocarcinomas.

Minor components of micropapillary and solid subtypes in lung invasive adenocarcinoma (≤ 3 cm): PET/CT findings and correlations with lymph node metastasis.

OBJECTIVE: To investigate the PET/CT findings in lung invasive adenocarcinoma with minor components of micropapillary or solid contents and its association with lymph node metastasis. MATERIALS AND METHODS: A total of 506 lung invasive adenocarcinoma (≤ 3 cm) patients who underwent a PET/CT examination and resection surgery were included. According to the proportion of solid/micropapillary components, the patients were classified into three groups: solid/micropapillary-negative (SMPN) (n = 258), solid/micropapillary-minor (SMPM; > 5% not predominant) (n = 158) and solid/micropapillary-predominant (SMPP; > 5% most dominant) (n = 90). The patients' PET/CT findings, including SUVmax, MTV, TLG and CT characteristics, and other clinical factors were compared by one-way ANOVA test. Logistic regression analysis was done to identify the most predictive findings for lymph node metastasis. RESULTS: The value of SUVmax, MTV, TLG and tumor size was highest in SMPP group, followed by SMPM and SMPN group (P < 0.001).The areas under the curve for SUVmax, MTV and TLG for node metastasis were 0.822, 0.843 and 0.835, respectively. Univariate analysis found that the SMPP and SMPM group had more lymph node metastasis than the SMPN group (P < 0.001). Furthermore, the lymph node metastasis group had higher CEA, SUVmax, MTV, TLG, tumor size and more pleural invasion (P < 0.001). Logistic regression analysis found that SMPP pathological type, SMPM pathological type, higher CEA and male patients were risk factors for lymph node metastasis (P < 0.01). CONCLUSIONS: Lung invasive adenocarcinoma with micropapillary or solid contents had higher SUVmax, MTV, TLG and tumor size and was associated with lymph node metastasis, even if they were not predominant.

ROS1 positive non-small cell lung cancer with pulmonary embolism in a 22-year woman.

ROS1-rearrangement occurs in 1-2% of non-small cell lung cancer (NSCLC). This mutation is predominantly seen in relatively young, non-smoker, female with adenocarcinoma. Association of pulmonary embolism with ROS1-rearranged NSCLC has been suggested. We report a case of a 22-year-old woman with ROS1-positive NSCLC and pulmonary embolism. This case possibly represents the youngest patient in the literature.

A qualitative transcriptional signature for the histological reclassification of lung squamous cell carcinomas and adenocarcinomas.

BACKGROUND: Targeted therapy for non-small cell lung cancer is histology dependent. However, histological classification by routine pathological assessment with hematoxylin-eosin staining and immunostaining for poorly differentiated tumors, particularly those from small biopsies, is still challenging. Additionally, the effectiveness of immunomarkers is limited by technical inconsistencies of immunostaining and lack of standardization for staining interpretation. RESULTS: Using gene expression profiles of pathologically-determined lung adenocarcinomas and squamous cell carcinomas, denoted as pADC and pSCC respectively, we developed a qualitative transcriptional signature, based on the within-sample relative gene expression orderings (REOs) of gene pairs, to distinguish ADC from SCC. The signature consists of two genes, KRT5 and AGR2, which has the stable REO pattern of KRT5 > AGR2 in pSCC and KRT5 < AGR2 in pADC. In the two test datasets with relative unambiguous NSCLC types, the apparent accuracy of the signature were 94.44 and 98.41%, respectively. In the other integrated dataset for frozen tissues, the signature reclassified 4.22% of the 805 pADC patients as SCC and 12% of the 125 pSCC patients as ADC. Similar results were observed in the clinical challenging cases, including FFPE specimens, mixed tumors, small biopsy specimens and poorly differentiated specimens. The survival analyses showed that the pADC patients reclassified as SCC had significantly shorter overall survival than the signature-confirmed pADC patients (log-rank p = 0.0123, HR = 1.89), consisting with the knowledge that SCC patients suffer poor prognoses than ADC patients. The proliferative activity, subtype-specific marker genes and consensus clustering analyses also supported the correctness of our signature. CONCLUSIONS: The non-subjective qualitative REOs signature could effectively distinguish ADC from SCC, which would be an auxiliary test for the pathological assessment of the ambiguous cases.

Reproducibility of visual estimation of lung adenocarcinoma subtype proportions.

In 2011, a new classification of lung adenocarcinoma was published. Since then there have been multiple studies regarding observer variability in predominant subtype determination, with levels of agreement generally in the weak to moderate range. In the 2011 and subsequent WHO 2015 classification, a recommendation was also made to visually assess and record the percentage of each subtype in 5% increments. The present study was initiated to determine the reproducibility of such gestalt assessments and to compare these data to a formal morphometric assessment. Five experienced pathologists reviewed multiple single images of 25 adenocarcinomas, taken at 2× and 10×, and estimated the percentage of lepidic, acinar, papillary, micropapillary, and solid components in 5% increments. After 2 months all the pathologists again reviewed the same images presented to them in a different order. We found that there was poor reproducibility within observers at 2× power using a 5% evaluation, but that this improved using 10% or 25% cutoffs. Use of 10× magnification allowed weak to moderate reproducibility at 5% increments, and this was again improved using 10% or 25% cutoffs. Correlation with morphometric assessment was poor except for the papillary and micropapillary subtypes. Differences among pathologists were generally low except for the acinar and, to a lesser degree, lepidic subtypes, which showed a wide spread of data. When estimating tumor subtype proportions, use of a 10× objective, and utilization of 10% or preferably 25% cutoffs provides a greater degree of consistency than a 5% cutoff.

Problems in the reproducibility of classification of small lung adenocarcinoma: an international interobserver study.

AIMS: The 2015 WHO classification for lung adenocarcinoma (ACA) provides criteria for adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (INV), but differentiating these entities can be difficult. As our understanding of prognostic significance increases, inconsistent classification is problematic. This study assesses agreement within an international panel of lung pathologists and identifies factors contributing to inconsistent classification. METHODS AND RESULTS: Sixty slides of small lung ACAs were reviewed digitally by six lung pathologists in three rounds, with consensus conferences and examination of elastic stains in round 3. The panel independently reviewed each case to assess final diagnosis, invasive component size and predominant pattern. The kappa value for AIS and MIA versus INV decreased from 0.44 (round 1) to 0.30 and 0.34 (rounds 2 and 3). Interobserver agreement for invasion (AIS versus other) decreased from 0.34 (round 1) to 0.29 and 0.29 (rounds 2 and 3). The range of the measured invasive component in a single case was up to 19.2 mm among observers. Agreement was excellent in tumours with high-grade cytology and fair with low-grade cytology. CONCLUSIONS: Interobserver agreement in small lung ACAs was fair to moderate, and improved minimally with elastic stains. Poor agreement is primarily attributable to subjectivity in pattern recognition, but high-grade cytology increases agreement. More reliable methods to differentiate histological patterns may be necessary, including refinement of the definitions as well as recognition of other features (such as high-grade cytology) as a formal part of routine assessment.

Cytomorphological criteria for separation of pulmonary adenocarcinomas from squamous cell carcinomas: A statistical learning approach.

BACKGROUND: Current therapy requires separation of non-small cell carcinomas into adenocarcinomas (AC) and squamous cell carcinomas (SCC). A meta-analysis has shown a pooled diagnostic sensitivity of 63% and specificity of 95% for the diagnosis of AC. While a number of cytomorphological features have been proposed for separation of AC from SCC, we are unaware of a statistically based analysis of cytomorphological features useful for separation of these two carcinomas. We performed logistic regression analysis of cytological features useful in classifying SCC and AC. DESIGN: Sixty-one Papanicolaou-stained fine needle aspiration specimens (29 AC/32 SCC) were reviewed by two board-certified cytopathologists for nine features (eccentric nucleoli, vesicular chromatin, prominent nucleoli, vacuolated cytoplasm, 3-dimensional cell balls, dark non-transparent chromatin, central nucleoli, single malignant cells and spindle-shaped cells). All cytological specimens had surgical biopsy results. Inter-rater agreement was assessed by Cohen's κ. Association between features and AC was determined using hierarchical logistic regression model where feature scores were nested within reviewers. A model to classify cases as SCC or AC was developed and verified by k-fold verification (k = 5). Classification performance was assessed using the area under the receiver operating characteristic curve. RESULTS: Observed rater agreement for scored features ranged from 49% to 82%. Kappa scores were clustered in three groups. Raters demonstrated good agreement for prominent nucleoli, vesicular chromatin and eccentric nuclei. Fair agreement was seen for 3-dimensional cell balls, dark non-transparent chromatin, and presence of spindle-shaped cells. Association of features with adenocarcinoma showed four statistically significant associations (P < 0.001) with adenocarcinoma. These features were prominent nucleoli, vesicular chromatin, eccentric nuclei and three-dimensional cell balls. Spindle-shaped cells and dark non-transparent chromatin were negatively associated with adenocarcinoma. CONCLUSIONS: Logistic regression analysis demonstrated six features helpful in separation of AC from SCC. Prominent nucleoli, vesicular chromatin, cell balls and eccentric nucleoli were positively associated with AC and demonstrated a P value of 0.001 or less. The presence of dark, non-transparent chromatin and spindle-shaped cells favoured the diagnosis of SCC.

Molecular Subtyping and Prognostic Assessment Based on Tumor Mutation Burden in Patients with Lung Adenocarcinomas.

The distinct molecular subtypes of lung cancer are defined by monogenic biomarkers, such as EGFR, KRAS, and ALK rearrangement. Tumor mutation burden (TMB) is a potential biomarker for response to immunotherapy, which is one of the measures for genomic instability. The molecular subtyping based on TMB has not been well characterized in lung adenocarcinomas in the Chinese population. Here we performed molecular subtyping based on TMB with the published whole exome sequencing data of 101 lung adenocarcinomas and compared the different features of the classified subtypes, including clinical features, somatic driver genes, and mutational signatures. We found that patients with lower TMB have a longer disease-free survival, and higher TMB is associated with smoking and aging. Analysis of somatic driver genes and mutational signatures demonstrates a significant association between somatic RYR2 mutations and the subtype with higher TMB. Molecular subtyping based on TMB is a potential prognostic marker for lung adenocarcinoma. Signature 4 and the mutation of RYR2 are highlighted in the TMB-High group. The mutation of RYR2 is a significant biomarker associated with high TMB in lung adenocarcinoma.

Pulmonary large cell neuroendocrine carcinoma with adenocarcinoma-like features: napsin A expression and genomic alterations.

Pulmonary large cell neuroendocrine carcinoma (LCNEC) is a highly aggressive malignancy, which was recently found to comprise three major genomic subsets: small cell carcinoma-like, non-small cell carcinoma (predominantly adenocarcinoma)-like, and carcinoid-like. To further characterize adenocarcinoma-like subset, here we analyzed the expression of exocrine marker napsin A, along with TTF-1, in a large series of LCNECs (n=112), and performed detailed clinicopathologic and genomic analysis of napsin A-positive cases. For comparison, we analyzed napsin A expression in other lung neuroendocrine neoplasms (177 carcinoids, 37 small cell carcinomas) and 60 lung adenocarcinomas. We found that napsin A was expressed in 15% of LCNEC (17/112), whereas all carcinoids and small cell carcinomas were consistently negative. Napsin A reactivity in LCNEC was focal in 12/17 cases, and weak or moderate in intensity in all cases, which was significantly lower in the extent and intensity than seen in adenocarcinomas (P<0.0001). The combination of TTF-1-diffuse/napsin A-negative or focal was typical of LCNEC but was rare in adenocarcinoma, and could thus serve as a helpful diagnostic clue. The diagnosis of napsin A-positive LCNECs was confirmed by classic morphology, diffuse labeling for at least one neuroendocrine marker, most consistently synaptophysin, and the lack of distinct adenocarcinoma component. Genomic analysis of 14 napsin A-positive LCNECs revealed the presence of mutations typical of lung adenocarcinoma (KRAS and/or STK11) in 11 cases. In conclusion, LCNECs are unique among lung neuroendocrine neoplasms in that some of these tumors exhibit low-level expression of exocrine marker napsin A, and harbor genomic alterations typical of adenocarcinoma. Despite the apparent close biological relationship, designation of adeno-like LCNEC as a separate entity from adenocarcinoma is supported by their distinctive morphology, typically diffuse expression of neuroendocrine marker(s) and aggressive behavior. Further studies are warranted to assess the clinical utility and optimal method of identifying adenocarcinoma-like and other subsets of LCNEC in routine practice.

Adenocarcinoma predominant pattern subtyping and nuclear grading in cytology: Is there a role in prognostication of advanced pulmonary adenocarcinomas?

INTRODUCTION: Primary lung adenocarcinomas (ADs) show varied architectural patterns, and pattern-based subtyping of ADs is currently recommended due to prognostic implications. Predicting AD patterns on cytology is challenging; however, cytological nuclear features appear to correlate with histological grade and survival in early stage lung ADs. The feasibility and value of AD pattern prediction and nuclear grading on cytology in advanced lung ADs is not known. We aimed to predict patterns and analyse nuclear features on cytology and evaluate their role in prognostication. METHODS: One-hundred patients of Stage III/IV lung AD with available matched cytology and histology samples were included. Cyto-patterns based on cell arrangement patterns (flat sheets vs three-dimensional clusters vs papillae) and cyto-nuclear score based on nuclear features (size, shape, contour), nucleoli (macronucleoli vs prominent vs inconspicuous), and nuclear chromatin were determined, and correlated with predominant histological-pattern observed on the matched small biopsy and outcome. RESULTS: Higher cyto-nuclear scores were observed with high-grade histo-patterns (solid, micropapillary and cribriform), while the predicted cyto-patterns did not correspond to the predominant pattern on histology in 77% cases. Highest cyto-histo agreement was observed for solid pattern (72%). High grade histo-patterns and cyto-nuclear scores > 3 showed a trend towards inferior survival (not significant). CONCLUSIONS: Nuclear grade scoring on cytology is simple to perform, and is predictive of high grade patterns. Its inclusion in routine reporting of cytology samples of lung ADs may be valuable.

[The 2015 classification of lung adenocarcinomas: Reproducibility in a Tunisian department specialised in thoracic pathology]. / La nouvelle classification 2015 des adénocarcinomes pulmonaires : reproductibilité au sein d'un service Tunisien spécialisé en pathologie thoracique.

INTRODUCTION: The new classification of lung cancer contains modifications of terminology and a new subdivision of the tumors with the most relevant modifications concerning the group of adenocarcinomas. The latter has been increasing and represents nowadays the most frequent type. Our aim was to assess the reproducibility of the new classification through the experience of a Department of Pathology specialized in thoracic pathology. METHODS: Our study included initially 106 cases diagnosed as adenocarcinomas and reviewed by 2 pathologists and 1 referee. Five cases were ruled out because they corresponded to squamous carcinoma according to the immunohistochemical findings. The same number of slides was reviewed without a limit of time. Statistical analysis was performed using the SPSS software. The Kappa index was estimated and a second coefficient: rho was analyzed. RESULTS: A total concordance was noticed in 82 cases (81.2%) and a discordance was noticed in 19 cases (18.8%). The agreement degree was good with an index Kappa estimated to 0.743 and a rho index reaching 0.763. CONCLUSION: Our study highlights the good reproducibility of the 2015 WHO classification of lung cancer among a trained team. Whereas, in order to improve the reproducibility of such a classification, even in non specialised departments, a training of the pathologists is necessary in order to highlight the prognostic impact of this classification.

Subclassification of lung adenocarcinoma through comprehensive multi-omics data to benefit survival outcomes.

OBJECTIVES: Lung adenocarcinoma (LUAD) is the most common subtype of non-small cell lung cancer. Understanding the molecular mechanisms underlying tumor progression is of great clinical significance. This study aims to identify novel molecular markers associated with LUAD subtypes, with the goal of improving the precision of LUAD subtype classification. Additionally, optimization efforts are directed towards enhancing insights from the perspective of patient survival analysis. MATERIALS AND METHODS: We propose an innovative feature-selection approach that focuses on LUAD classification, which is comprehensive and robust. The proposed method integrates multi-omics data from The Cancer Genome Atlas (TCGA) and leverages a synergistic combination of max-relevance and min-redundancy, least absolute shrinkage and selection operator, and Boruta algorithms. These selected features were deployed in six machine-learning classifiers: logistic regression, random forest, support vector machine, naive Bayes, k-Nearest Neighbor, and XGBoost. RESULTS: The proposed approach achieved an area under the receiver operating characteristic curve (AUC) of 0.9958 for LR. Notably, the accuracy and AUC of a composite model incorporating copy number, methylation, as well as RNA- sequencing data for expression of exons, genes, and miRNA mature strands surpassed the accuracy and AUC metrics of models with single-omics data or other multi-omics combinations. Survival analyses, revealed the SVM classifier to elicit optimal classification, outperforming that achieved by TCGA. To enhance model interpretability, SHapley Additive exPlanations (SHAP) values were utilized to elucidate the impact of each feature on the predictions. Gene Ontology (GO) enrichment analysis identified significant biological processes, molecular functions, and cellular components associated with LUAD subtypes. CONCLUSION: In summary, our feature selection process, based on TCGA multi-omics data and combined with multiple machine learning classifiers, proficiently identifies molecular subtypes of lung adenocarcinoma and their corresponding significant genes. Our method could enhance the early detection and diagnosis of LUAD, expedite the development of targeted therapies and, ultimately, lengthen patient survival.

A dual data stream hybrid neural network for classifying pathological images of lung adenocarcinoma.

Lung cancer has seriously threatened human health due to its high lethality and morbidity. Lung adenocarcinoma, in particular, is one of the most common subtypes of lung cancer. Pathological diagnosis is regarded as the gold standard for cancer diagnosis. However, the traditional manual screening of lung cancer pathology images is time consuming and error prone. Computer-aided diagnostic systems have emerged to solve this problem. Current research methods are unable to fully exploit the beneficial features inherent within patches, and they are characterized by high model complexity and significant computational effort. In this study, a deep learning framework called Multi-Scale Network (MSNet) is proposed for the automatic detection of lung adenocarcinoma pathology images. MSNet is designed to efficiently harness the valuable features within data patches, while simultaneously reducing model complexity, computational demands, and storage space requirements. The MSNet framework employs a dual data stream input method. In this input method, MSNet combines Swin Transformer and MLP-Mixer models to address global information between patches and the local information within each patch. Subsequently, MSNet uses the Multilayer Perceptron (MLP) module to fuse local and global features and perform classification to output the final detection results. In addition, a dataset of lung adenocarcinoma pathology images containing three categories is created for training and testing the MSNet framework. Experimental results show that the diagnostic accuracy of MSNet for lung adenocarcinoma pathology images is 96.55 %. In summary, MSNet has high classification performance and shows effectiveness and potential in the classification of lung adenocarcinoma pathology images.