Invasive size is an independent predictor of survival in pulmonary adenocarcinoma

Genomic and immune heterogeneity of multiple synchronous lung adenocarcinoma at different developmental stages

Multiple synchronous lung cancers (MSLCs) constitute a unique subtype of lung cancer. To explore the genomic and immune heterogeneity across different pathological stages of MSLCs, we analyse 16 MSLCs from 8 patients using single-cell RNA-seq, single-cell TCR sequencing, and bulk whole-exome sequencing. Our investigation indicates clonally independent tumours with convergent evolution driven by shared driver mutations. However, tumours from the same individual exhibit few shared mutations, indicating independent origins. During the transition from pre-invasive to invasive adenocarcinoma, we observe a shift in T cell phenotypes characterized by increased Treg cells and exhausted CD8+ T cells, accompanied by diminished cytotoxicity. Additionally, invasive adenocarcinomas exhibit greater neoantigen abundance and a more diverse TCR repertoire, indicating heightened heterogeneity. In summary, despite having a common genetic background and environmental exposure, our study emphasizes the individuality of MSLCs at different stages, highlighting their unique genomic and immune characteristics.

Multimodal fusion of liquid biopsy and CT enhances differential diagnosis of early-stage lung adenocarcinoma

This research explores the potential of multimodal fusion for the differential diagnosis of early-stage lung adenocarcinoma (LUAD) (tumor sizes < 2 cm). It combines liquid biopsy biomarkers, specifically extracellular vesicle long RNA (evlRNA) and the computed tomography (CT) attributes. The fusion model achieves an impressive area under receiver operating characteristic curve (AUC) of 91.9% for the four-classification of adenocarcinoma, along with a benign-malignant AUC of 94.8% (sensitivity: 89.1%, specificity: 94.3%). These outcomes outperform the diagnostic capabilities of the single-modal models and human experts. A comprehensive SHapley Additive exPlanations (SHAP) is provided to offer deep insights into model predictions. Our findings reveal the complementary interplay between evlRNA and image-based characteristics, underscoring the significance of integrating diverse modalities in diagnosing early-stage LUAD.

Accuracy of solid portion size measured on multiplanar volume rendering images for assessing invasiveness in lung adenocarcinoma manifesting as subsolid nodules

Background

The solid component of subsolid nodules (SSNs) is closely associated with the invasiveness of lung adenocarcinoma, and its accurate assessment is crucial for selecting treatment method. Therefore, this study aimed to evaluate the accuracy of solid component size within SSNs measured on multiplanar volume rendering (MPVR) and compare it with the dimensions of invasive components on pathology.

Methods

A pilot study was conducted using a chest phantom to determine the optimal MPVR threshold for the solid component within SSN, and then clinical validation was carried out by retrospective inclusion of patients with pathologically confirmed solitary SSN from October 2020 to October 2021. The radiological tumor size on MPVR and solid component size on MPVR (RSSm) and on lung window (RSSl) were measured. The size of the tumor and invasion were measured on the pathological section, and the invasion, fibrosis, and inflammation within SSNs were also recorded. The measurement difference between computed tomography (CT) and pathology, inter-observer and inter-measurement agreement were analyzed. Receiver operating characteristic (ROC) analysis and Bland-Altman plot were performed to evaluate the diagnostic efficiency of MPVR.

Results

A total of 142 patients (mean age, 54±11 years, 39 men) were retrospectively enrolled in the clinical study, with 26 adenocarcinomas in situ, 92 minimally invasive adenocarcinomas (MIAs), and 24 invasive adenocarcinomas (IAs). The RSSl was significantly smaller than pathological invasion size with fair inter-measurement agreement [intraclass correlation coefficient (ICC) =0.562, P<0.001] and moderate interobserver agreement (ICC =0.761, P<0.001). The RSSm was significantly larger than pathological invasion size with the excellent inter-measurement agreement (ICC =0.829, P<0.001) and excellent (ICC =0.952, P<0.001) interobserver agreement. ROC analysis showed that the cutoff value of RSSm for differentiating adenocarcinoma in situ from MIA and MIA from IA was 1.85 and 6.45 mm (sensitivity: 93.8% and 95.5%, specificity: 85.7% and 88.2%, 95% confidence internal: 0.914-0.993 and 0.900-0.983), respectively. The positive predictive value-and negative predictive value of MPVR in predicting invasiveness were 92.8% and 100%, respectively.

Conclusions

Using MPVR to predict the invasive degree of SSN had high accuracy and good inter-observer agreement, which is superior to lung window measurements and helpful for clinical decision-making.

Ex-vivo 1.5T MR Imaging versus CT in Estimating the Size of the Pathologically Invasive Component of Lung Adenocarcinoma Spectrum Lesions

PURPOSE

The purpose of this study was to investigate whether ex-vivo MRI enables accurate estimation of the invasive component of lung adenocarcinoma.

METHODS

We retrospectively reviewed 32 patients with lung adenocarcinoma who underwent lung lobectomy. The specimens underwent MRI at 1.5T. The boundary between the lesion and the normal lung was evaluated on a 5-point scale in each three MRI sequences, and a one-way analysis of variance and post-hoc tests were performed. The invasive component size was measured histopathologically. The maximum diameter of each solid component measured on CT and MR T1-weighted (T1W) images and the maximum size obtained from histopathologic images were compared using the Wilcoxon signed-rank test. Inter-reader agreement was evaluated using intraclass correlation coefficients (ICC).

RESULTS

T1W images were determined to be optimal for the delineation of the lesions (P < 0.001). The histopathologic invasive area corresponded to the area where the T1W ex-vivo MR image showed a high signal intensity that was almost equal to the intravascular blood signal. The maximum diameter of the solid component on CT was overestimated compared with the maximum invasive size on histopathology (mean, 153%; P < 0.05), while that on MRI was evaluated mostly accurately without overestimation (mean, 108%; P = 0.48). The interobserver reliability of the measurements using CT and MRI was good (ICC = 0.71 on CT, 0.74 on MRI).

CONCLUSION

Ex-vivo MRI was more accurate than conventional CT in delineating the invasive component of lung adenocarcinoma.

Combined model integrating deep learning, radiomics, and clinical data to classify lung nodules at chest CT

OBJECTIVES

The study aimed to develop a combined model that integrates deep learning (DL), radiomics, and clinical data to classify lung nodules into benign or malignant categories, and to further classify lung nodules into different pathological subtypes and Lung Imaging Reporting and Data System (Lung-RADS) scores.

MATERIALS AND METHODS

The proposed model was trained, validated, and tested using three datasets: one public dataset, the Lung Nodule Analysis 2016 (LUNA16) Grand challenge dataset (n = 1004), and two private datasets, the Lung Nodule Received Operation (LNOP) dataset (n = 1027) and the Lung Nodule in Health Examination (LNHE) dataset (n = 1525). The proposed model used a stacked ensemble model by employing a machine learning (ML) approach with an AutoGluon-Tabular classifier. The input variables were modified 3D convolutional neural network (CNN) features, radiomics features, and clinical features. Three classification tasks were performed: Task 1: Classification of lung nodules into benign or malignant in the LUNA16 dataset; Task 2: Classification of lung nodules into different pathological subtypes; and Task 3: Classification of Lung-RADS score. Classification performance was determined based on accuracy, recall, precision, and F1-score. Ten-fold cross-validation was applied to each task.

RESULTS

The proposed model achieved high accuracy in classifying lung nodules into benign or malignant categories in LUNA 16 with an accuracy of 92.8%, as well as in classifying lung nodules into different pathological subtypes with an F1-score of 75.5% and Lung-RADS scores with an F1-score of 80.4%.

CONCLUSION

Our proposed model provides an accurate classification of lung nodules based on the benign/malignant, different pathological subtypes, and Lung-RADS system.

An oncogenic CTNNB1 mutation is predictive of post-operative recurrence-free survival in an EGFR-mutant lung adenocarcinoma

The Wnt/β-catenin pathway is known to be frequently dysregulated in various human malignancies. Alterations in the genes encoding the components of Wnt/β-catenin pathway have also been described in lung adenocarcinoma. Notably however, the clinical impacts of Wnt/β-catenin pathway alterations in lung adenocarcinoma have not been fully evaluated to date. We here investigated the prognostic implications of single gene variations in 174 cases of surgically resected lung adenocarcinoma tested using targeted next-generation sequencing. Screening of the prognostic impact of single gene alterations identified an association between CTNNB1 mutation and poor recurrence-free survival in EGFR-mutant LUADs. Based on these results, the entire cohort was stratified into three groups in accordance with the mutational status of Wnt/β-catenin pathway genes (i.e. oncogenic CTNNB1 mutation [CTNNB1-ONC], other Wnt/β-catenin pathway gene mutations [Wnt/β-catenin-OTHER], and wild type for Wnt/β-catenin pathway genes [Wnt/β-catenin-WT]). The clinicopathologic characteristics and survival outcomes of these groups were then compared. Oncogenic CTNNB1 and other Wnt/β-catenin pathway gene mutations were identified in 10 (5.7%) and 14 cases (8.0%), respectively. The CTNNB1-ONC group cases displayed histopathologic features of conventional non-mucinous adenocarcinoma with no significant differences from those of the other groups. Using β-catenin immunohistochemistry, we found that the CTNNB1-ONC group displayed aberrant nuclear staining more frequently, but only in 60% of the samples. The LUADs harboring a CTNNB1-ONC exhibited significantly poorer RFS outcomes than the other groups, regardless of the β-catenin IHC status. This was a pronounced finding in the EGFR-mutant LUADs only in subgroup analysis, which was then confirmed by multivariate analysis. Nevertheless, no significant OS differences between these Wnt/β-catenin groups were evident. Hence, oncogenic CTNNB1 mutations may be found in about 6% of lung adenocarcinomas and may predict post-operative recurrence in EGFR-mutant LUADs. Aberrant nuclear β-catenin staining on IHC appears to be insufficient as a surrogate marker of an oncogenic CTNNB1 mutation.

Adding predictive and diagnostic values of pulmonary ground-glass nodules on lung cancer via novel non-invasive tests

Pulmonary ground-glass nodules (GGNs) are highly associated with lung cancer. Extensive studies using thin-section high-resolution CT images have been conducted to analyze characteristics of different types of GGNs in order to evaluate and determine the predictive and diagnostic values of GGNs on lung cancer. Accurate prediction of their malignancy and invasiveness is critical for developing individualized therapies and follow-up strategies for a better clinical outcome. Through reviewing the recent 5-year research on the association between pulmonary GGNs and lung cancer, we focused on the radiologic and pathological characteristics of different types of GGNs, pointed out the risk factors associated with malignancy, discussed recent genetic analysis and biomarker studies (including autoantibodies, cell-free miRNAs, cell-free DNA, and DNA methylation) for developing novel diagnostic tools. Based on current progress in this research area, we summarized a process from screening, diagnosis to follow-up of GGNs.

Fusion of CT images and clinical variables based on deep learning for predicting invasiveness risk of stage I lung adenocarcinoma

PURPOSE

To develop a novel multimodal data fusion model by incorporating computed tomography (CT) images and clinical variables based on deep learning for predicting the invasiveness risk of stage I lung adenocarcinoma that manifests as ground-glass nodules (GGNs), and compare the diagnostic performance of it with that of radiologists.

METHODS

A total of 1946 patients with solitary and histopathologically confirmed GGNs with maximum diameter less than 3 cm were retrospectively enrolled. The training dataset containing 1704 GGNs was augmented by resampling, scaling, random cropping, etc., to generate new training data. A multimodal data fusion model based on residual learning architecture and two multilayer perceptron with attention mechanism combining CT images with patient general data and serum tumor markers was built. The distance-based confidence scores (DCS) were calculated and compared among multimodal data models with different combinations. An observer study was conducted and the prediction performance of the fusion algorithms was compared with that of the two radiologists by an independent testing dataset with 242 GGNs.

RESULTS

Among the whole GGNs, 606 GGNs are confirmed as invasive adenocarcinoma (IA) and 1340 are non-IA. The proposed novel multimodal data fusion model combining CT images, patient general data and serum tumor markers achieved the highest accuracy (88.5%), Area under a ROC curve (AUC) (0.957), F1 (81.5%), F1_weighted (81.9%) and Matthews correlation coefficient (MCC) (73.2%) for classifying between IA and non-IA GGNs, which was even better than the senior radiologist's performance (accuracy, 86.1%). In addition, the DCSs for multimodal data suggested that CT image had a stronger influence (0.9540) quantitatively than general data (0.6726) or tumor marker (0.6971).

CONCLUSION

This study demonstrated that the feasibility of integrating different types of data including CT images and clinical variables, and the multimodal data fusion model yielded higher performance for distinguishing IA from non-IA GGNs. This article is protected by copyright. All rights reserved.

A comparative study for the evaluation of CT-based conventional, radiomic, combined conventional and radiomic, and delta-radiomic features, and the prediction of the invasiveness of lung adenocarcinoma manifesting as ground-glass nodules

AIM

To investigate and compare the performance of conventional, radiomic, combined, and delta-radiomic features to predict the invasiveness of lung adenocarcinoma manifesting as ground-glass nodules (GGNs).

MATERIALS AND METHODS

The present retrospective study included 216 GGNs confirmed surgically as pulmonary adenocarcinomas. All the thin-section computed tomography (CT) images were imported into the software of the United Imaging Intelligence research portal, and radiomic features were extracted with three-dimensional (3D) regions of interest. Least Absolute Shrinkage and Selection Operator was used to select the optimal radiomic features. Four models were constructed, including conventional, radiomic, combined conventional and radiomic, and delta-radiomic models. The receiver operating characteristic curves were built to evaluate the validity of these.

RESULTS

The type, long diameter, shape, margin, vacuole, air bronchus, vascular convergence, and pleural traction exhibited significant differences between pre-invasive lesions (PILs)/minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IA) groups were selected for conventional model building. Nine radiomic features were selected to build the radiomic model. The four models indicated optimal performance (AUC > 0.7). The radiomic and combined models exhibited the highest diagnostic efficiency, and their AUC were 0.89 and 0.88 in the training set, and 0.87 and 0.88 in the validation set, respectively. The delta-radiomic model indicated that the AUC was 0.83 in the training set, and 0.76 in the validation set. Finally, the conventional model exhibited an AUC in the training and validation sets of 0.78 and 0.76.

CONCLUSIONS

The radiomic model and combined model, in particular, and the delta-radiomic model all demonstrated improved diagnostic efficiency in differentiating IA from PIL/MIA than that of the conventional model.

Increased Tumor Intrinsic Growth Potential and Decreased Immune Function Orchestrate the Progression of Lung Adenocarcinoma

Background

The overall 5-year survival of lung cancer was reported to be only ~15%, with lung adenocarcinoma (LUAD) as the main pathological subtype. Before developing into invasive stages, LUAD undergoes pre-invasive stages of adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA), where surgical resection gives an excellent 5-year survival rate. Given the dramatic decline of prognosis from pre-invasive to invasive stages, a deeper understanding of key molecular changes driving the progression of LUAD is highly needed.

Methods

In this study, we performed whole-exome sequencing and RNA sequencing on surgically resected 24 AIS, 74 MIA, 99 LUAD specimens, and their adjacent paired normal tissues. Survival data were obtained by follow-up after surgery. Key molecular events were found by comparing the gene expression profiles of tumors with different stages. Finally, to measure the level of imbalance between tumor intrinsic growth potential and immune microenvironment, a tumor progressive (TP) index was developed to predict tumor progression and patients’ survival outcome and validated by external datasets.

Results

As tumors progressed to more invasive stages, they acquired higher growth potential, mutational frequency of tumor suppressor genes, somatic copy number alterations, and tumor mutation burden, along with suppressed immune function. To better predict tumor progression and patients’ outcome, TP index were built to measure the imbalance between tumor intrinsic growth potential and immune microenvironment. Patients with a higher TP index had significantly worse recurrence-free survival [Hazard ratio (HR), 10.47; 95% CI, 3.21–34.14; p < 0.0001] and overall survival (OS) [Hazard ratio (HR), 4.83e8; 95% CI, 0–Inf; p = 0.0013]. We used The Cancer Genome Atlas (TCGA)-LUAD dataset for validation and found that patients with a higher TP index had significantly worse OS (HR, 1.10; 95% CI, 0.83–1.45; p = 0.048), demonstrating the prognostic value of the TP index for patients with LUAD.

Conclusions

The imbalance of tumor intrinsic growth potential and immune function orchestrate the progression of LUAD, which can be measured by TP index. Our study provided new insights into predicting survival of patients with LUAD and new target discovery for LUAD through assessing the imbalance between tumor intrinsic growth potential and immune function.

Development and validation of a preoperative CT-based radiomic nomogram to predict pathology invasiveness in patients with a solitary pulmonary nodule: a machine learning approach, multicenter, diagnostic study

OBJECTIVES

To develop and validate a preoperative CT-based nomogram combined with radiomic and clinical-radiological signatures to distinguish preinvasive lesions from pulmonary invasive lesions.

METHODS

This was a retrospective, diagnostic study conducted from August 1, 2018, to May 1, 2020, at three centers. Patients with a solitary pulmonary nodule were enrolled in the GDPH center and were divided into two groups (7:3) randomly: development (n = 149) and internal validation (n = 54). The SYSMH center and the ZSLC Center formed an external validation cohort of 170 patients. The least absolute shrinkage and selection operator (LASSO) algorithm and logistic regression analysis were used to feature signatures and transform them into models.

RESULTS

The study comprised 373 individuals from three independent centers (female: 225/373, 60.3%; median [IQR] age, 57.0 [48.0-65.0] years). The AUCs for the combined radiomic signature selected from the nodular area and the perinodular area were 0.93, 0.91, and 0.90 in the three cohorts. The nomogram combining the clinical and combined radiomic signatures could accurately predict interstitial invasion in patients with a solitary pulmonary nodule (AUC, 0.94, 0.90, 0.92) in the three cohorts, respectively. The radiomic nomogram outperformed any clinical or radiomic signature in terms of clinical predictive abilities, according to a decision curve analysis and the Akaike information criteria.

CONCLUSIONS

This study demonstrated that a nomogram constructed by identified clinical-radiological signatures and combined radiomic signatures has the potential to precisely predict pathology invasiveness.

KEY POINTS

• The radiomic signature from the perinodular area has the potential to predict pathology invasiveness of the solitary pulmonary nodule. • The new radiomic nomogram was useful in clinical decision-making associated with personalized surgical intervention and therapeutic regimen selection in patients with early-stage non-small-cell lung cancer.

Updates in grading and invasion assessment in lung adenocarcinoma

The pathologic evaluation of lung adenocarcinoma, because of greater understanding of disease progression and prognosis, has become more complex. It is clear that histologic growth patterns reflect indolent and aggressive disease, resulting in clearer morphologic groups that can be the underpinning of a grading system. In addition, the progression of adenocarcinoma from a tumor that preserves alveolar architecture to one that remodels and effaces lung structure has led to criteria that reflect invasive rather than in-situ growth. While some of these are based on tumor cell growth pattern, aspects of this remodeling from desmoplasia to artifacts of lung collapse and sectioning, can lead to difficult to interpret patterns with lower reproducibility between observers. Such scenarios are examined to provide updates on new histologic concepts and to highlight ongoing problem areas.

DeepRePath: Identifying the Prognostic Features of Early-Stage Lung Adenocarcinoma Using Multi-Scale Pathology Images and Deep Convolutional Neural Networks

Simple Summary

Pathology images are vital for understanding solid cancers. In this study, we created DeepRePath using multi-scale pathology images with two-channel deep learning to predict the prognosis of patients with early-stage lung adenocarcinoma (LUAD). DeepRePath demonstrated that it could predict the recurrence of early-stage LUAD with average area under the curve scores of 0.77 and 0.76 in cohort I and cohort II (external validation set), respectively. Pathological features found to be associated with a high probability of recurrence included tumor necrosis, discohesive tumor cells, and atypical nuclei. In conclusion, DeepRePath can improve the treatment modality for patients with early-stage LUAD through recurrence prediction.

Abstract

The prognosis of patients with lung adenocarcinoma (LUAD), especially early-stage LUAD, is dependent on clinicopathological features. However, its predictive utility is limited. In this study, we developed and trained a DeepRePath model based on a deep convolutional neural network (CNN) using multi-scale pathology images to predict the prognosis of patients with early-stage LUAD. DeepRePath was pre-trained with 1067 hematoxylin and eosin-stained whole-slide images of LUAD from the Cancer Genome Atlas. DeepRePath was further trained and validated using two separate CNNs and multi-scale pathology images of 393 resected lung cancer specimens from patients with stage I and II LUAD. Of the 393 patients, 95 patients developed recurrence after surgical resection. The DeepRePath model showed average area under the curve (AUC) scores of 0.77 and 0.76 in cohort I and cohort II (external validation set), respectively. Owing to low performance, DeepRePath cannot be used as an automated tool in a clinical setting. When gradient-weighted class activation mapping was used, DeepRePath indicated the association between atypical nuclei, discohesive tumor cells, and tumor necrosis in pathology images showing recurrence. Despite the limitations associated with a relatively small number of patients, the DeepRePath model based on CNNs with transfer learning could predict recurrence after the curative resection of early-stage LUAD using multi-scale pathology images.

A clinicopathological study of lung adenocarcinomas with pure ground-glass opacity > 3 cm on high-resolution computed tomography

OBJECTIVES

This study aimed to discuss whether a diameter of 3 cm is a threshold for diagnosing lung adenocarcinomas presenting with radiological pure ground-glass mass (PGGM, pure ground-glass opacity > 3 cm) as adenocarcinomas in situ or minimally invasive adenocarcinomas (AIS-MIAs). Another aim was to identify CT features and patient prognosis that differentiate AIS-MIAs from invasive adenocarcinomas (IACs) in patients with PGGMs.

METHODS

From June 2007 to October 2015, 69 resected PGGMs with HRCT and followed up for ≥ 5 years were included in this study and divided into AIS-MIA (n = 13) and IAC (n = 56) groups. Firth's logistic regression model was performed to determine CT characteristics that helped distinguish IACs from AIS-MIAs. The discriminatory power of the significant predictors was tested with the area under the receiver operating characteristics curve (AUC). Disease recurrence was also evaluated.

RESULTS

Univariable and multivariable analyses identified that the mean CT attenuation (odds ratio: 1.054, p = 0.0087) was the sole significant predictor for preoperatively discriminating IACs from AIS-MIAs in patients with PGGMs. The CT attenuation had an excellent differentiating accuracy (AUC: 0.981), with the optimal cut-off value at -600 HU (sensitivity: 87.5%; specificity: 100%). Additionally, no recurrence was observed in patients manifesting with PGGMs > 3 cm, and the 5-year recurrence-free survival and overall survival rates were both 100%, even in cases of IAC.

CONCLUSIONS

This study demonstrated that PGGMs > 3 cm could still be AIS-MIAs. When PGGMs are encountered in clinical practice, the CT value may be the only valuable parameter to preoperatively distinguish IACs from AIS-MIAs.

KEY POINTS

• Patients with pure ground-glass opacity > 3 cm in diameter are rare but can be diagnosed as adenocarcinomas in situ or minimally invasive adenocarcinomas. • The mean CT attenuation is the sole significant CT parameter that differentiates invasive adenocarcinoma from adenocarcinoma in situ or minimally invasive adenocarcinoma in patients with pure ground-glass opacity > 3 cm. • Lung adenocarcinoma with pure ground-glass opacity > 3 cm has an excellent prognosis, even in cases of invasive adenocarcinoma.

3D deep learning based classification of pulmonary ground glass opacity nodules with automatic segmentation

Classifying ground-glass lung nodules (GGNs) into atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC) on diagnostic CT images is important to evaluate the therapy options for lung cancer patients. In this paper, we propose a joint deep learning model where the segmentation can better facilitate the classification of pulmonary GGNs. Based on our observation that masking the nodule to train the model results in better lesion classification, we propose to build a cascade architecture with both segmentation and classification networks. The segmentation model works as a trainable preprocessing module to provide the classification-guided 'attention' weight map to the raw CT data to achieve better diagnosis performance. We evaluate our proposed model and compare with other baseline models for 4 clinically significant nodule classification tasks, defined by a combination of pathology types, using 4 classification metrics: Accuracy, Average F1 Score, Matthews Correlation Coefficient (MCC), and Area Under the Receiver Operating Characteristic Curve (AUC). Experimental results show that the proposed method outperforms other baseline models on all the diagnostic classification tasks.

The value of percentile base on computed tomography histogram in differentiating the invasiveness of adenocarcinoma appearing as pure ground-glass nodules

To investigate the value of percentile base on computed tomography (CT) histogram analysis for distinguishing invasive adenocarcinoma (IA) from adenocarcinoma in situ (AIS) or micro invasive adenocarcinoma (MIA) appearing as pure ground-glass nodules.A total of 42 cases of pure ground-glass nodules that were surgically resected and pathologically confirmed as lung adenocarcinoma between January 2015 and May 2019 were included. Cases were divided into IA and AIS/MIA in the study. The percentile on CT histogram was compared between the 2 groups. Univariate and multivariate logistic regression were used to determine which factors demonstrated a significant effect on invasiveness. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) was used to evaluate the predictive ability of individual characteristics and the combined model.The 4 histogram parameters (25th percentile, 55th percentile, 95th percentile, 97.5th percentile) and the combined model all showed a certain diagnostic value. The combined model demonstrated the best diagnostic performance. The AUC values were as follows: 25th percentile = 0.693, 55th percentile = 0.706, 95th percentile = 0.713, 97.5th percentile = 0.710, and combined model = 0.837 (all P < .05).The percentile of histogram parameters help to improve the ability to radiologically determine the invasiveness of lung adenocarcinoma appearing as pure ground-glass nodules.

Diagnostic performance for pulmonary adenocarcinoma on CT: comparison of radiologists with and without three-dimensional convolutional neural network

OBJECTIVES

To compare diagnostic performance for pulmonary invasive adenocarcinoma among radiologists with and without three-dimensional convolutional neural network (3D-CNN).

METHODS

Enrolled were 285 patients with adenocarcinoma in situ (AIS, n = 75), minimally invasive adenocarcinoma (MIA, n = 58), and invasive adenocarcinoma (IVA, n = 152). A 3D-CNN model was constructed with seven convolution-pooling and two max-pooling layers and fully connected layers, in which batch normalization, residual connection, and global average pooling were used. Only the flipping process was performed for augmentation. The output layer comprised two nodes for two conditions (AIS/MIA and IVA) according to prognosis. Diagnostic performance of the 3D-CNN model in 285 patients was calculated using nested 10-fold cross-validation. In 90 of 285 patients, results from each radiologist (R1, R2, and R3; with 9, 14, and 26 years of experience, respectively) with and without the 3D-CNN model were statistically compared.

RESULTS

Without the 3D-CNN model, accuracy, sensitivity, and specificity of the radiologists were as follows: R1, 70.0%, 52.1%, and 90.5%; R2, 72.2%, 75%, and 69%; and R3, 74.4%, 89.6%, and 57.1%, respectively. With the 3D-CNN model, accuracy, sensitivity, and specificity of the radiologists were as follows: R1, 72.2%, 77.1%, and 66.7%; R2, 74.4%, 85.4%, and 61.9%; and R3, 74.4%, 93.8%, and 52.4%, respectively. Diagnostic performance of each radiologist with and without the 3D-CNN model had no significant difference (p > 0.88), but the accuracy of R1 and R2 was significantly higher with than without the 3D-CNN model (p < 0.01).

CONCLUSIONS

The 3D-CNN model can support a less-experienced radiologist to improve diagnostic accuracy for pulmonary invasive adenocarcinoma without deteriorating any diagnostic performances.

KEY POINTS

• The 3D-CNN model is a non-invasive method for predicting pulmonary invasive adenocarcinoma in CT images with high sensitivity. • Diagnostic accuracy by a less-experienced radiologist was better with the 3D-CNN model than without the model.

Lung Adenocarcinoma at CT with 0.25-mm Section Thickness and a 2048 Matrix: High-Spatial-Resolution Imaging for Predicting Invasiveness

Background High-spatial-resolution (HSR) CT provides detailed information and clear delineation of lung anatomy and disease states. HSR CT may have high diagnostic performance for predicting invasiveness of lung adenocarcinoma. Purpose To examine the diagnostic performance of HSR CT in predicting the invasiveness of lung adenocarcinoma. Materials and Methods In this retrospective study, 89 consecutive patients with adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IVA) were included who underwent surgery for lung cancer between January 2018 and December 2019. All patients underwent HSR CT with 0.25-mm section thickness and a 2048 matrix. Two independent observers evaluated the images for the presence or absence of the following HSR CT findings: lobulation, spiculation, pleural indentation, vessel convergence, homogeneity of ground-glass opacity, reticulation, irregularity and centrality of solid portion, and air bronchiologram (irregularity, disruption, or dilatation). The total diameter (≤1.6 cm or >1.6 cm) and the longest diameter of the solid portion (≤0.8 cm or >0.8 cm) were evaluated. Logistic regression models were used to identify findings associated with MIA plus IVA. Receiver operating characteristic analysis was performed to determine diagnostic performance. Results Eighty-nine patients (mean, 69 years ± 11 [standard deviation]; 49 men) were evaluated. The size of the nodules with invasion was a mean of 2.5 cm ± 1.2. Univariable analysis revealed lobulation, spiculation, pleural indentation, irregular and central solid portion, air bronchiologram with disruption and/or irregular dilatation, and total and solid portion diameters as associated with MIA plus IVA (all, P < .05). After adjustment for age, sex, and pack-years of smoking, disruption of air bronchogram and solid portion diameter greater than 0.8 cm remained as predictors of invasiveness (P = .001 and P = .02, respectively). The diagnostic performance of these two findings combined were as follows: sensitivity of 97% (59 of 61 patients; 95% confidence interval: 94%, 100%) and specificity of 86% (19 of 22 patients; 95% confidence interval: 65%, 97%), with an area under the curve of 0.94. Conclusion Using high-spatial-resolution CT, disruption of air bronchiologram and a solid portion greater than 0.8 cm were independently associated with a greater likelihood of invasiveness in lung adenocarcinoma. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Lynch and Oh in this issue.

Factors distinguishing invasive from pre-invasive adenocarcinoma presenting as pure ground glass pulmonary nodules

Background

To investigate predictors of pathological invasiveness and prognosis of lung adenocarcinoma in patients with pure ground-glass nodules (pGGNs).

Methods

Clinical and computed tomography (CT) features of invasive adenocarcinomas (IACs) and pre-IACs were retrospectively compared in 641 consecutive patients with pGGNs and confirmed lung adenocarcinomas who had undergone postoperative CT follow-up. Potential predictors of prognosis were investigated in these patients.

Results

Of 659 pGGNs in 641 patients, 258 (39.1%) were adenocarcinomas in situ, 265 (40.2%) were minimally invasive adenocarcinomas, and 136 (20.6%) were IACs. Respective optimal cutoffs for age, serum carcinoembryonic antigen concentration, maximal diameter, mean diameter, and CT density for distinguishing pre-IACs from IACs were 53 years, 2.19 ng/mL, 10.78 mm, 10.09 mm, and − 582.28 Hounsfield units (HU). Univariable analysis indicated that sex, age, maximal diameter, mean diameter, CT density, and spiculation were significant predictors of lung IAC. In multivariable analysis age, maximal diameter, and CT density were significant predictors of lung IAC. During a median follow-up of 41 months no pGGN IACs recurred.

Conclusions

pGGNs may be lung IACs, especially in patients aged > 55 years with lesions that are > 1 cm in diameter and exhibit CT density > − 600 HU. pGGN IACs of < 3 cm in diameter have good post-resection prognoses.

Radiological-pathological correlation of subsolid pulmonary nodules: A single centre retrospective evaluation of the 2011 IASLC adenocarcinoma classification system

INTRODUCTION

The 2011 IASLC classification system proposes guidelines for radiologists and pathologists to classify adenocarcinomas spectrum lesions as preinvasive, minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IA). IA portends the worst clinical prognosis, and the imaging distinction between MIA and IA is controversial.

MATERIALS AND METHODS

Subsolid pulmonary nodules resected by microcoil localization over a three-year period were retrospectively reviewed by three chest radiologists and a pulmonary pathologist. Nodules were classified radiologically based on preoperative computed tomography (CT), with the solid nodule component measured on mediastinal windows applied to high-frequency lung kernel reconstructions, and pathologically according to 2011 IASLC criteria. Radiology interobserver and radiological-pathological variability of nodule classification, and potential reasons for nodule classification discordance were assessed.

RESULTS

Seventy-one subsolid nodules in 67 patients were included. The average size of invasive disease focus at histopathology was 5 mm (standard deviation 5 mm). Radiology interobserver agreement of nodule classification was good (Cohen's Kappa = 0.604, 95 % CI: 0.447 to 0.761). Agreement between consensus radiological interpretation and pathological category was fair (Cohen's Kappa = 0.236, 95 % CI: 0.054-0.421). Radiological and pathological nodule classification were concordant in 52 % (37 of 71) of nodules. The IASLC proposed CT solid component cut-off of 5 mm to distinguish MIA and IA yielded a sensitivity of 59 % and specificity of 80 %. Common reasons for nodule classification discordance included multiple solid components within a nodule on CT, scar and stromal collapse at pathology, and measurement variability.

CONCLUSION

Solid component(s) within persistent part-solid pulmonary nodules raise suspicion for invasive adenocarcinoma. Preoperative imaging classification is frequently discordant from final pathology, reflecting interpretive and technical challenges in radiological and pathological analysis.

Combination of generative adversarial network and convolutional neural network for automatic subcentimeter pulmonary adenocarcinoma classification

Background

The efficient and accurate diagnosis of pulmonary adenocarcinoma before surgery is of considerable significance to clinicians. Although computed tomography (CT) examinations are widely used in practice, it is still challenging and time-consuming for radiologists to distinguish between different types of subcentimeter pulmonary nodules. Although there have been many deep learning algorithms proposed, their performance largely depends on vast amounts of data, which is difficult to collect in the medical imaging area. Therefore, we propose an automatic classification system for subcentimeter pulmonary adenocarcinoma, combining a convolutional neural network (CNN) and a generative adversarial network (GAN) to optimize clinical decision-making and to provide small dataset algorithm design ideas.

Methods

A total of 206 nodules with postoperative pathological labels were analyzed. Among them were 30 adenocarcinomas in situ (AISs), 119 minimally invasive adenocarcinomas (MIAs), and 57 invasive adenocarcinomas (IACs). Our system consisted of two parts, a GAN-based image synthesis, and a CNN classification. First, several popular existing GAN techniques were employed to augment the datasets, and comprehensive experiments were conducted to evaluate the quality of the GAN synthesis. Additionally, our classification system processes were based on two-dimensional (2D) nodule-centered CT patches without the need of manual labeling information.

Results

For GAN-based image synthesis, the visual Turing test showed that even radiologists could not tell the GAN-synthesized from the raw images (accuracy: primary radiologist 56%, senior radiologist 65%). For CNN classification, our progressive growing wGAN improved the performance of CNN most effectively (area under the curve =0.83). The experiments indicated that the proposed GAN augmentation method improved the classification accuracy by 23.5% (from 37.0% to 60.5%) and 7.3% (from 53.2% to 60.5%) in comparison with training methods using raw and common augmented images respectively. The performance of this combined GAN and CNN method (accuracy: 60.5%±2.6%) was comparable to the state-of-the-art methods, and our CNN was also more lightweight.

Conclusions

The experiments revealed that GAN synthesis techniques could effectively alleviate the problem of insufficient data in medical imaging. The proposed GAN plus CNN framework can be generalized for use in building other computer-aided detection (CADx) algorithms and thus assist in diagnosis.

A Bayesian hidden Potts mixture model for analyzing lung cancer pathology images

Digital pathology imaging of tumor tissues, which captures histological details in high resolution, is fast becoming a routine clinical procedure. Recent developments in deep-learning methods have enabled the identification, characterization, and classification of individual cells from pathology images analysis at a large scale. This creates new opportunities to study the spatial patterns of and interactions among different types of cells. Reliable statistical approaches to modeling such spatial patterns and interactions can provide insight into tumor progression and shed light on the biological mechanisms of cancer. In this article, we consider the problem of modeling a pathology image with irregular locations of three different types of cells: lymphocyte, stromal, and tumor cells. We propose a novel Bayesian hierarchical model, which incorporates a hidden Potts model to project the irregularly distributed cells to a square lattice and a Markov random field prior model to identify regions in a heterogeneous pathology image. The model allows us to quantify the interactions between different types of cells, some of which are clinically meaningful. We use Markov chain Monte Carlo sampling techniques, combined with a double Metropolis-Hastings algorithm, in order to simulate samples approximately from a distribution with an intractable normalizing constant. The proposed model was applied to the pathology images of $205$ lung cancer patients from the National Lung Screening trial, and the results show that the interaction strength between tumor and stromal cells predicts patient prognosis (P = $0.005$). This statistical methodology provides a new perspective for understanding the role of cell-cell interactions in cancer progression.

Optimal method for measuring invasive size that predicts survival in invasive mucinous adenocarcinoma of the lung

PURPOSE

The purpose of this study was to determine the optimal method for measuring pathological invasive size that predicts prognosis in invasive mucinous adenocarcinoma (IMA).

METHODS

We analyzed patients who underwent complete surgical resection for lung IMA. The invasive size of IMA was measured using two methods: (1) excluding lepidic method (ELM), that is, lepidic component was excluded from the invasive area regardless of alveolar mucin and (2) including lepidic method (ILM), that is, lepidic component was included as invasive area if alveolar space was filled with mucin. The prognostic predictability of ELM and ILM on survival was assessed using univariable and multivariable Cox regression models. The discriminative power was assessed using concordance probability estimate (CPE) and Akaike's information criteria (AIC), and the prognostic impact of the newly redefined pathological stage according to ELM or ILM was also assessed.

RESULTS

A total of 101 patients were included. The median invasive size via ELM and ILM was 1.4 cm (range, 0.0-7.7 cm) and 2.1 cm (range, 0.0-14.2 cm), respectively. ELM had better discriminative power than ILM (ELM, HR = 1.38, AIC = 110.19, CPE = 0.671; ILM, HR = 1.19, AIC = 111.52, CPE = 0.655). Although the survival curves based on ILM crossed between T3 and T4, the overall survival (OS) curves based on ELM were sufficiently distinct from one another.

CONCLUSIONS

ELM has higher discriminative power for OS, and thus the optimal method for measuring the pathological invasive size of IMA should exclude the lepidic component regardless of alveolar mucin.

Pulmonary adenocarcinoma appearing as ground-glass opacity nodules identified using non-enhanced and contrast-enhanced CT texture analysis: A retrospective analysis

The present study aimed to investigate the ability of CT-based texture analysis to differentiate invasive adenocarcinoma (IA) from pre-invasive lesions (PIL) or minimally IA (MIA) appearing as ground-glass opacity (GGO) nodules, and to further compare the performance of non-enhanced CT (NECT) images with that of contrast-enhanced CT (CECT) images. A total of 77 patients with GGO nodules and surgically confirmed pulmonary adenocarcinoma were included in the present retrospective study. Each GGO nodule was manually segmented and its texture features were extracted from NECT and CECT images using in-house developed software coded in MATLAB (MathWorks). The independent-samples t-test was used to select the texture features with statistically significant differences between IA and MIA/PIL. Multivariate logistic regression and receiver operating characteristics (ROC) curve analyses were performed to identify predictive features. Of the 77 GGO nodules, 12 were atypical adenomatous hyperplasia or adenocarcinoma in situ (15.6%), 36 were MIA (46.8%) and 29 were IA (37.7%). IA and MIA/PIL exhibited significant differences in most histogram features and gray-level co-occurrence matrix features (P<0.05). Multivariate logistic regression and ROC curve analyses revealed that smaller energy and higher entropy were significant differentiators of IA from MIA and PIL, irrespective of whether NECT images [area under the curve (AUC): 0.839, 0.859] or CECT images (AUC: 0.818, 0.820) are used. Texture analysis of CT images, regardless of whether NECT or CECT is used, has the potential to distinguish IA from PIL or MIA, particularly the parameters of energy and entropy. Furthermore, NECT images were simpler to obtain and no contrast agent was required; thus, analysis with NECT may be a preferred choice.

Marginal radiomics features as imaging biomarkers for pathological invasion in lung adenocarcinoma

OBJECTIVES

Lung adenocarcinomas which manifest as ground-glass nodules (GGNs) have different degrees of pathological invasion and differentiating among them is critical for treatment. Our goal was to evaluate the addition of marginal features to a baseline radiomics model on computed tomography (CT) images to predict the degree of pathologic invasiveness.

METHODS

We identified 236 patients from two cohorts (training, n = 189; validation, n = 47) who underwent surgery for GGNs. All GGNs were pathologically confirmed as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IA). The regions of interest were semi-automatically annotated and 40 radiomics features were computed. We selected features using L1-norm regularization to build the baseline radiomics model. Additional marginal features were developed using the cumulative distribution function (CDF) of intratumoral intensities. An improved model was built combining the baseline model with CDF features. Three classifiers were tested for both models.

RESULTS

The baseline radiomics model included five features and resulted in an average area under the curve (AUC) of 0.8419 (training) and 0.9142 (validation) for the three classifiers. The second model, with the additional marginal features, resulted in AUCs of 0.8560 (training) and 0.9581 (validation). All three classifiers performed better with the added features. The support vector machine showed the most performance improvement (AUC improvement = 0.0790) and the best performance was achieved by the logistic classifier (validation AUC = 0.9825).

CONCLUSION

Our novel marginal features, when combined with a baseline radiomics model, can help differentiate IA from AIS and MIA on preoperative CT scans.

KEY POINTS

• Our novel marginal features could improve the existing radiomics model to predict the degree of pathologic invasiveness in lung adenocarcinoma.

Atypical goblet cell hyperplasia occurs in CPAM 1, 2, and 3, and is a probable precursor lesion for childhood adenocarcinoma

Congenital pulmonary airway malformation (CPAM) is a developmental disorder. Types 1-2-3 are the more common ones. Atypical goblet cell hyperplasia (AGCH) in CPAM might be a precursor lesion for pulmonary adenocarcinomas. In nine out of 33 CPAM cases, types 1-3 showed foci of goblet cell proliferations. As these cells completely replace normal epithelium, we prefer to name these proliferations AGCH. In 5 cases, adenocarcinomas were seen (AC). All cases were analyzed for proteins possibly being associated with CPAM development: fibroblast growth factor 10 (FGF10) and receptor 2 (FGFR2), forkhead box A1 (FOXA1) and A2 (FOXA2), MUC protein 5AC (MUC5AC), human epidermal growth factor receptor 2 (erbB2, HER2/neu), hepatocyte nuclear factor 4α (HNF4α), SOX2, and Ying Yang protein 1 (YY1). By next generation sequencing, AGCH and adenocarcinomas were evaluated for driver mutations. Expression for FGF10, FGFR2, FOXA1, and FOXA2 was seen in CPAM epithelium and stroma, but not differently in AGCH and AC. SOX2 was positive in CPAM epithelium and AGCH, however weakly in AC. YY1 and MUC5AC showed more intense staining in AGCH and AC than in CPAM epithelium. HER2 was intensely expressed in AC and less intensely in AGCH, but not in CPAM epithelium. KRAS mutation in exon 2 was detected in all AGCH and AC, but was absent in CPAM epithelia. AGCH can arise in CPAM types 1-3. Oncogenic KRAS mutation seems to be the oncogenic driver already in AGCH, proving its role as a precursor lesion for adenocarcinoma. It might upregulate HER2 at the protein level. YY1 seems to be involved in carcinogenesis.

Lung Cancer Staging

Lung cancer staging is a foundation of patient care, informing management decisions and prognosis. This comprehensive overview of the current 8th edition American Joint Committee on Cancer Cancer Staging Manual addresses common difficulties in staging, such as measuring the invasive component of adenocarcinomas and staging multiple lung nodules.

Genomic and immune profiling of pre-invasive lung adenocarcinoma

Adenocarcinoma in situ and minimally invasive adenocarcinoma are the pre-invasive forms of lung adenocarcinoma. The genomic and immune profiles of these lesions are poorly understood. Here we report exome and transcriptome sequencing of 98 lung adenocarcinoma precursor lesions and 99 invasive adenocarcinomas. We have identified EGFR, RBM10, BRAF, ERBB2, TP53, KRAS, MAP2K1 and MET as significantly mutated genes in the pre/minimally invasive group. Classes of genome alterations that increase in frequency during the progression to malignancy are revealed. These include mutations in TP53, arm-level copy number alterations, and HLA loss of heterozygosity. Immune infiltration is correlated with copy number alterations of chromosome arm 6p, suggesting a link between arm-level events and the tumor immune environment.

A BAYESIAN MARK INTERACTION MODEL FOR ANALYSIS OF TUMOR PATHOLOGY IMAGES

With the advance of imaging technology, digital pathology imaging of tumor tissue slides is becoming a routine clinical procedure for cancer diagnosis. This process produces massive imaging data that capture histological details in high resolution. Recent developments in deep-learning methods have enabled us to identify and classify individual cells from digital pathology images at large scale. Reliable statistical approaches to model the spatial pattern of cells can provide new insight into tumor progression and shed light on the biological mechanisms of cancer. We consider the problem of modeling spatial correlations among three commonly seen cells observed in tumor pathology images. A novel geostatistical marking model with interpretable underlying parameters is proposed in a Bayesian framework. We use auxiliary variable MCMC algorithms to sample from the posterior distribution with an intractable normalizing constant. We demonstrate how this model-based analysis can lead to sharper inferences than ordinary exploratory analyses, by means of application to three benchmark datasets and a case study on the pathology images of 188 lung cancer patients. The case study shows that the spatial correlation between tumor and stromal cells predicts patient prognosis. This statistical methodology not only presents a new model for characterizing spatial correlations in a multitype spatial point pattern conditioning on the locations of the points, but also provides a new perspective for understanding the role of cell-cell interactions in cancer progression.

Computed Tomography-Based Score Indicative of Lung Cancer Aggression (SILA) Predicts the Degree of Histologic Tissue Invasion and Patient Survival in Lung Adenocarcinoma Spectrum

OBJECTIVE

Most computed tomography (CT)-detected lung cancers are adenocarcinomas (ACs), representing lesions with variable tissue invasion, aggressiveness, and clinical outcome. Visual radiologic characterization of AC pulmonary nodules is both inconsistent and inadequate to confidently predict histopathologic classification or prognosis. Comprehensive pathologic interpretation requires full nodule resection. We have described a computerized scoring system for AC detected on CT scans that can noninvasively estimate the degree of histologic invasion and simultaneously predict patient survival.

METHODS

The Computer-Aided Nodule Assessment and Risk Yield has been validated to characterize CT-detected nodules across the spectrum of AC. With the use of unsupervised clustering, nine natural exemplars were identified as basic radiographic features of AC nodules. We now introduce the Score Indicative of Lung Cancer Aggression (SILA), which is a cumulative aggregate of normalized distributions of ordered Computer-Aided Nodule Assessment and Risk Yield exemplars. The SILA values for each of 237 unique nodules in AC were compared with the histopathologically defined maximum linear extent of tumor invasion. With use of the SILA, Kaplan-Meier survival and Cox proportionality analysis were performed on patients with stage I AC, who comprised a subset of our cohort.

RESULTS

The SILA discriminated between indolent and invasive AC (p < 0.0001). In addition, prediction of linear extent of histopathologic tumor invasion was possible. In stage I AC, three separate SILA prognosis groups were identified: indolent, intermediate, and poor, with 5-year survival rates of 100%, 79%, 58%, respectively. Cox proportionality hazard modeling predicted a 50% increase in mortality, for a 0.1 unit increase in the SILA over a median follow-up time of 3.6 years (p < 0.0002).

CONCLUSIONS

The SILA is a computer-based analytic measure allowing noninvasive approximation of histologic invasion and prediction of patient survival in CT-detected AC nodules.

Peripheral blood telomere alterations in ground glass opacity (GGO) lesions may suggest malignancy

A ground glass opacity (GGO) lung lesion may represent early stage adenocarcinoma, which has an excellent prognosis upon prompt surgical resection. However, GGO lesions have broad differential diagnoses, including both benign and malignant lesions. Our objective was to study telomere length and telomerase activity in patients with suspected lung cancer in which GGO was the predominant radiographic feature. Knowledge of telomere biology may help distinguish malignant from benign radiographic lesions and guide risk assessment of these lesions. Peripheral blood samples were taken from 22 patients with suspected adenocarcinoma with the GGO radiographic presentation. Multidisciplinary discussion confirmed the need for surgery in all cases. We used an age and gender‐matched group without known lung disease as a control. Telomere length and aggregates were assessed by quantitative fluorescence in situ hybridization (QFISH) and quantitative PCR. Cell senescence was evaluated by senescence‐associated heterochromatin foci. Subjects with GGO lesions had a higher percentage of lymphocytes with shorter telomeres (Q‐FISH, P = 0.003). Furthermore, relative telomere length was also reduced among the GGO cases (qPCR, P < 0.05). Increased senescence was observed in the GGO group compared to controls (P < 0.001), with significant correlation between the senescence‐associated heterochromatin foci and aggregate formation (r = −0.7 and r = −0.44 for cases and controls, respectively). In conclusion, patients with resectable early adenocarcinoma demonstrate abnormal telomere length and cell senescence in peripheral blood leukocytes compared to control subjects. Abnormal telomere biology in the peripheral blood may increase suspicion of early adenocarcinoma among patients with GGO lesions.

Looking for Lepidic Component inside Invasive Adenocarcinomas Appearing as CT Solid Solitary Pulmonary Nodules (SPNs): CT Morpho-Densitometric Features and 18-FDG PET Findings

Objective

To investigate CT morphologic and densitometric features and 18-FDG PET findings of surgically excised lung adenocarcinomas "mixed subtype" with predominant lepidic component, appearing as solid solitary pulmonary nodules (SPNs) on CT scan.

Materials and Methods

Approval for this study was given from each local institutional review board according to its retrospective nature. Nodules pathologically classified as lung adenocarcinoma mixed subtype with bronchioloalveolar otherwise lepidic predominant component, in three different Italian institutions (Napoli; Varese; Parma), were retrospectively selected.

Results

22 patients were identified. The number of SPNs with smooth margins was significantly lower with respect to the number of SPNs with spiculated margins (p: 0.033), radiating spiculations (p: 0.019), and notch sign (p: 0.011). Mean contrast enhancement (CE) was 53.34 HU (min 5.5 HU, max 112 HU); considering 15 HU as cut-off value, CE was positive in 20/22 cases. No significant correlation was found between size and CE. Mean SUVmax was 2.21, ranging from 0.2 up to 7.5 units; considering 2.5 units as cut-off, SUVmax was positive in 7/22 cases. The number of SPNs with positive CE was significantly higher than the number of SPNs with positive SUVmax (p: 0.0005).

Conclusion

CT generally helps in identifying solid SPN suspicious for malignancy but 18-FDG PET may result in false-negative evaluation; when 18-FDG PET findings of a solid SPN are negative even though CT morphology and CE suggest malignancy, radiologist should consider that lepidic component may be present inside the invasive tumor, despite the absence of ground glass.

Implications of the Eighth Edition of the TNM Proposal: Invasive Versus Total Tumor Size for the T Descriptor in Pathologic Stage I-IIA Lung Adenocarcinoma

INTRODUCTION

The eighth edition of the TNM staging system included the proposal that the T descriptor be determined according to the invasive component, excluding lepidic component, for nonmucinous lung adenocarcinomas. We sought to conduct a clinicopathologic comparative analysis of the newly proposed classification using invasive size versus total tumor size.

METHODS

Patients who underwent lung resection for primary lung adenocarcinoma with pathologic stage (p-Stage) I-IIA (based on total size [t]) were reviewed (n = 1704). Pathologic invasive size was measured, and tumors were reclassified using invasive size (i). Cumulative incidence of recurrence and lung cancer-specific cumulative incidence of death were analyzed using a competing-risks approach. Prognostic discrimination by p-Stage(t) and p-Stage(i) was evaluated using a concordance index (C-index).

RESULTS

The use of invasive size resulted in downstaging in 377 of 1704 patients (22%), with twice as many patients with p-Stage IA1 (IA1[i] versus IA1[t]: 389 [23%] versus 195 [11%]). However, outcomes were similar between the two groups (IA1[i] versus IA1[t]: 5-year cumulative incidence of recurrence, 11% versus 13%; 5-year lung cancer-specific cumulative incidence of death, 5% versus 7%). Prognostic discrimination by p-Stage(i) was better than by p-Stage(t) (C-index for p-Stage[i] versus p-Stage[t]: recurrence, 0.614 versus 0.593; lung cancer-specific death, 0.634 versus 0.621).

CONCLUSIONS

When invasive size, rather than total size, was used for the T descriptor, a larger number of patients were classified with a favorable prognosis (p-Stage IA1) and better prognostic discrimination of p-Stage I-IIA nonmucinous lung adenocarcinomas was achieved.

3D Deep Learning from CT Scans Predicts Tumor Invasiveness of Subcentimeter Pulmonary Adenocarcinomas

: Identification of early-stage pulmonary adenocarcinomas before surgery, especially in cases of subcentimeter cancers, would be clinically important and could provide guidance to clinical decision making. In this study, we developed a deep learning system based on 3D convolutional neural networks and multitask learning, which automatically predicts tumor invasiveness, together with 3D nodule segmentation masks. The system processes a 3D nodule-centered patch of preprocessed CT and learns a deep representation of a given nodule without the need for any additional information. A dataset of 651 nodules with manually segmented voxel-wise masks and pathological labels of atypical adenomatous hyperplasia (AAH), adenocarcinomas in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive pulmonary adenocarcinoma (IA) was used in this study. We trained and validated our deep learning system on 523 nodules and tested its performance on 128 nodules. An observer study with 2 groups of radiologists, 2 senior and 2 junior, was also investigated. We merged AAH and AIS into one single category AAH-AIS, comprising a 3-category classification in our study. The proposed deep learning system achieved better classification performance than the radiologists; in terms of 3-class weighted average F1 score, the model achieved 63.3% while the radiologists achieved 55.6%, 56.6%, 54.3%, and 51.0%, respectively. These results suggest that deep learning methods improve the yield of discriminative results and hold promise in the CADx application domain, which could help doctors work efficiently and facilitate the application of precision medicine. SIGNIFICANCE: Machine learning tools are beginning to be implemented for clinical applications. This study represents an important milestone for this emerging technology, which could improve therapy selection for patients with lung cancer.

Predictive accuracy of lepidic growth subtypes in early-stage adenocarcinoma of the lung by quantitative CT histogram and FDG-PET

OBJECTIVES

The aim of this study was to analyze the accuracy of computed tomography (CT) and F-18 fluorodeoxyglucose-positron emission tomography/CT (FDG-PET/CT) to distinguish lepidic growth adenocarcinoma (LGA), including adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and lepidic-predominant adenocarcinoma, all of which have favorable survival outcomes, from the more aggressive and invasive non-LGA subtypes.

MATERIALS AND METHODS

We identified 225 patients with c-0/I adenocarcinoma of the lung who underwent PET/CT and 3DCT followed by complete resection. Maximum standardized uptake values (SUVmax) of FDG and several histogram parameters were analyzed. Histological grades were classified according to the predominant subtype (G1: lepidic; G3: micropapillary or solid; and G2: subtypes other than G1/G3).

RESULTS

The proportion of pathological invasive factors (lymphatic vessel involvement/blood vessel invasion/pleural invasion/lymph node metastasis) of patients with preinvasive adenocarcinoma, G1, G2, and G3 tumors were 0%, 3.6%, 48.0%, and 100%, respectively; p <  0.001). Multivariate analysis with CT-related parameters demonstrated that 75^th percentile CT attenuation value (75^th%, p <  0.001) and maximum CT attenuation value (maxCT, p =  0.009) were associated with incidence of non-LGA, whereas the value of SUVmax demonstrated a significant correlation (p <  0.001). When all patients were dichotomized according to ground-glass opacities (GGO)/solid-dominancy for CT maximum diameter, a significant correlation with non-LGA was shown in patients with solid-dominant tumor on SUVmax (p <  0.001) and with GGO-dominant tumor on 75^th% (p =  0.006) and maxCT (p =  0.007). The combination of one of the two significant histogram parameters and SUVmax revealed higher predictive performance for pathological high malignant features (positive pathological invasive factors, non-LGA, and the highly malignant subtype covering G2 with moderately or poorly-differentiated carcinoma and G3) than the individual use of either factor.

CONCLUSION

The 75^th%, maxCT, and SUVmax were highly useful in distinguishing LGA from non-LGA in c-0/I adenocarcinoma.

Use of a Radiomics Model to Predict Tumor Invasiveness of Pulmonary Adenocarcinomas Appearing as Pulmonary Ground-Glass Nodules

Background

It is important to distinguish the classification of lung adenocarcinoma. A radiomics model was developed to predict tumor invasiveness using quantitative and qualitative features of pulmonary ground-glass nodules (GGNs) on chest CT.

Materials and Methods

A total of 599 GGNs [including 202 preinvasive lesions and 397 minimally invasive and invasive pulmonary adenocarcinomas (IPAs)] were evaluated using univariate, multivariate, and logistic regression analyses to construct a radiomics model that predicted invasiveness of GGNs. In primary cohort (comprised of patients scanned from August 2012 to July 2016), preinvasive lesions were distinguished from IPAs based on pure or mixed density (PM), lesion shape, lobulated border, and pleural retraction and 35 other quantitative parameters (P <0.05) using univariate analysis. Multivariate analysis showed that PM, lobulated border, pleural retraction, age, and fractal dimension (FD) were significantly different between preinvasive lesions and IPAs. After logistic regression analysis, PM and FD were used to develop a prediction nomogram. The validation cohort was comprised of patients scanned after Jan 2016.

Results

The model showed good discrimination between preinvasive lesions and IPAs with an area under curve (AUC) of 0.76 [95% CI: 0.71 to 0.80] in ROC curve for the primary cohort. The nomogram also demonstrated good discrimination in the validation cohort with an AUC of 0.79 [95% CI: 0.71 to 0.88].

Conclusions

For GGNs, PM, lobulated border, pleural retraction, age, and FD were features discriminating preinvasive lesions from IPAs. The radiomics model based upon PM and FD may predict the invasiveness of pulmonary adenocarcinomas appearing as GGNs.

Identification of genes associated with matrix metalloproteinases in invasive lung adenocarcinoma

The aim of the present study was to identify genes with similar function to that of matrix metalloproteinases (MMPs) in invasive lung adenocarcinoma (AC) and to screen the transcription factors that regulate MMPs. The gene expression dataset GSE2514, including 20 invasive lung AC samples and 19 adjacent normal lung samples, was downloaded from the Gene Expression Omnibus database. Differentially expressed genes (DEGs) were screened using the limma package in R. Genes with similar function to MMPs were identified by K-means clustering. Their correlations with MMPs were validated using Pearson correlation analysis. The expression of MMPs in lung cancer and normal tissues was evaluated by western blot analysis. Protein-protein interaction (PPI) network and transcriptional regulatory network analyses were performed with Retrieval of Interacting Genes and Database for Annotation, Visualization and Integrated Discovery, respectively. As a result, 269 DEGs were identified between invasive lung AC samples and normal lung samples, including 78 upregulated and 191 downregulated genes. Four MMPs (MMP1, MMP7, MMP9 and MMP12), which were upregulated in lung AC, were clustered into one group with other genes, including NAD(P)H quinone oxidoreductase 1, claudin 3 (CLDN3), S100 calcium-binding protein P, serine protease inhibitor Kazal type 1, collagen type XI α 1 chain, periostin and desmoplakin (DSP), following cluster analysis. Pearson correlation analysis further confirmed correlations between MMP9-CLDN3, MMP9-DSP and MMP12-DSP. PPI network analysis also indicated multiple interactions between MMPs-associated genes. Furthermore, MMPs were commonly regulated by CCAAT/enhancer binding protein α transcription factor. These findings may provide further insight into the mechanisms of MMPs in invasive lung AC.

Clinicopathological Characteristics and Mutations Driving Development of Early Lung Adenocarcinoma: Tumor Initiation and Progression

Lung cancer is the leading cause of cancer-related deaths worldwide, with lung adenocarcinoma representing the most common lung cancer subtype. Among all lung adenocarcinomas, the most prevalent subset develops via tumorigenesis and progression from atypical adenomatous hyperplasia (AAH) to adenocarcinoma in situ (AIS), to minimally invasive adenocarcinoma (MIA), to overt invasive adenocarcinoma with a lepidic pattern. This stepwise development is supported by the clinicopathological and molecular characteristics of these tumors. In the 2015 World Health Organization classification, AAH and AIS are both defined as preinvasive lesions, whereas MIA is identified as an early invasive adenocarcinoma that is not expected to recur if removed completely. Recent studies have examined the molecular features of lung adenocarcinoma tumorigenesis and progression. EGFR-mutated adenocarcinoma frequently develops via the multistep progression. Oncogene-induced senescence appears to decrease the frequency of the multistep progression in KRAS- or BRAF-mutated adenocarcinoma, whose tumor evolution may be associated with epigenetic alterations and kinase-inactive mutations. This review summarizes the current knowledge of tumorigenesis and tumor progression in early lung adenocarcinoma, with special focus on its clinicopathological characteristics and their associations with driver mutations (EGFR, KRAS, and BRAF) as well as on its molecular pathogenesis and progression.

Preoperative bronchoscopic cancer confirmation does not increase risk of recurrence in stage1A non-small cell lung cancer

OBJECTIVE

This study was conducted to evaluate the risk of recurrence possibly caused by preoperative bronchoscopic cancer confirmation in stage1A non-small cell lung cancer.

METHODS

One hundred and seventy-nine cases of peripheral non-small cell lung cancer (including 151 adenocarcinoma) with no more than 3 cm in their tumor longer diameter were selected. All patients underwent preoperative diagnostic bronchoscopy followed by lobectomy, and were demonstrated to have pathologically free of lymph node involvement and pleural involvement. Radiological and pathological low-grade adenocarcinomas were excluded. Of 179 cases, 95 were confirmed lung cancer by bronchoscope (Group 1) and rest 84 had failed cancer confirmation by bronchoscope before surgery (Group 2). Forty-eight pairs for non-small cell lung cancer and 41 pairs for adenocarcinoma were identified from each group by propensity caliper matching. Kaplan-Meier method and log-rank test were performed on matched groups, and Cox proportional hazard model analysis was performed on whole matched cases.

RESULTS

Log-rank test revealed no significant inferiority of recurrence-free survival of Group 1 in both all-NSCLC and adenocarcinoma subset. Cox proportional hazard model analysis also revealed that the 'presence of preoperative bronchoscopic cancer confirmation' dose not increase risk of recurrence in both NSCLC and adenocarcinoma subset.

CONCLUSIONS

It is unlikely that preoperative bronchoscopic cancer confirmation would increase recurrence risk in stage1A non-small cell lung cancer; however, a future prospective study with larger cohorts would be warranted to validate the results.

The ratio of cancer cells to stroma within the invasive area is a histologic prognostic parameter of lung adenocarcinoma

OBJECTIVES

This study evaluated whether the proportion of cancer cells to non-cancerous stroma within the invasive area is associated with the prognosis of patients with lung adenocarcinoma.

MATERIALS AND METHODS

A total of 127 patients with lung adenocarcinomas with tumors larger than 3 cm in total size were enrolled in this study. We classified the tumors according to the ratio of area occupied by cancer cells within the invasive area (Type A: more than 50% of the invasive area, Type B: 10-50%, and Type C: less than 10%) and analyzed the clinicopathological differences between Types A, B, and C.

RESULTS

The invasive size of Type A tumors (n = 35) was significantly larger than those of the other two tumor types; however, there was no significant difference in the invasive size between Types B (n = 65) and C (n = 25) tumors. The recurrence-free survival time of patients with Type C tumors was significantly longer than those of patients with Type A and B (P < .001) tumors. Multivariate analysis revealed that Type C tumor was an independent favorable prognostic factor (P = .037) but that invasive size was not. The invasive area of Type C tumor was composed of a significantly higher proportion of collapsed elastic fibers than the invasive areas of Type A and B tumors (P < .001).

CONCLUSION

A lower cancer cell to stroma ratio within the invasive area could be a significant prognostic factor in lung adenocarcinoma, suggesting that not only the invasive size but also the invasive character might be an important histologic prognostic parameter.

Radiomic analysis of pulmonary ground-glass opacity nodules for distinction of preinvasive lesions, invasive pulmonary adenocarcinoma and minimally invasive adenocarcinoma based on quantitative texture analysis of CT

Objective

To identify the differences among preinvasive lesions, minimally invasive adenocarcinomas (MIAs) and invasive pulmonary adenocarcinomas (IPAs) based on radiomic feature analysis with computed tomography (CT).

Methods

A total of 109 patients with ground-glass opacity lesions (GGOs) in the lungs determined by CT examinations were enrolled, all of whom had received a pathologic diagnosis. After the manual delineation and segmentation of the GGOs as regions of interest (ROIs), the patients were subdivided into three groups based on pathologic analyses: the preinvasive lesions (including atypical adenomatous hyperplasia and adenocarcinoma in situ) subgroup, the MIA subgroup and the IPA subgroup. Next, we obtained the texture features of the GGOs. The data analysis was aimed at finding both the differences between each pair of the groups and predictors to distinguish any two pathologic subtypes using logistic regression. Finally, a receiver operating characteristic (ROC) curve was applied to accurately evaluate the performances of the regression models.

Results

We found that the voxel count feature (P<0.001) could be used as a predictor for distinguishing IPAs from preinvasive lesions. However, the surface area feature (P=0.040) and the extruded surface area feature (P=0.013) could be predictors of IPAs compared with MIAs. In addition, the correlation feature (P=0.046) could distinguish preinvasive lesions from MIAs better.

Conclusions

Preinvasive lesions, MIAs and IPAs can be discriminated based on texture features within CT images, although the three diseases could all appear as GGOs on CT images. The diagnoses of these three diseases are very important for clinical surgery.

HRCT morphological characteristics distinguishing minimally invasive pulmonary adenocarcinoma from invasive pulmonary adenocarcinoma appearing as subsolid nodules with a diameter of ≤3 cm

AIM

To differentiate retrospectively the morphological characteristics at high-resolution computed tomography (CT) between minimally invasive pulmonary adenocarcinoma (MIA) and invasive pulmonary adenocarcinoma (IAC) appearing as subsolid nodules (SNs) with a diameter of ≤3 cm and to provide information to help operative decision-making.

MATERIALS AND METHODS

The patient notes of 260 patients with SNs of ≤3 cm in diameter (98 with MIA and 162 with IAC) confirmed at surgery and histopathology from September 2008 to June 2012 were reviewed retrospectively at the Department of Radiology, Weifang Respiratory Disease Hospital. Sixty-seven patients had pure ground-glass nodules (PGGNs) and 193 had mixed ground-glass nodules (MGGNs). Patients were grouped according to the final pathology: minimally invasive MIA and IAC. The HRCT characteristics were compared between the two groups.

RESULTS

There were statistically significant differences in the pattern, shape, diameter of solid components, proportion of solid components, CT radiodensity values of the ground-glass and solid components, borders, margins, air bronchograms, microvascular signs, and pleural indentations of the nodules between the two groups (all p<0.05). Multivariate and receiver operating characteristic (ROC) analyses indicated significant predictors of MIAs were as follows: small lesion diameter (≤14.7 mm), solid components ≤7 mm, <50% of solid components, low CT radiodensity values of the solid components (≤-107 HU), air bronchograms in the ground-glass opacity components, and microvascular signs.

CONCLUSION

The morphological characteristics at high-resolution CT can be used to differentiate between MIAs and IACs appearing as SNs with a diameter of ≤3 cm and provide information to help operative decision-making.

Radiologic-Pathologic Correlation of Solid Portions on Thin-section CT Images in Lung Adenocarcinoma: A Multicenter Study

BACKGROUND

Measuring the size of invasiveness on computed tomography (CT) for the T descriptor size was deemed important in the 8th edition of the TNM lung cancer classification. We aimed to correlate the maximal dimensions of the solid portions using both lung and mediastinal window settings on CT imaging with the pathologic invasiveness (> 0.5 cm) in lung adenocarcinoma patients.

MATERIALS AND METHODS

The study population consisted of 378 patients with a histologic diagnosis of adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), invasive adenocarcinoma (IVA)-lepidic, IVA-acinar and/or IVA-papillary, and IVA-micropapillary and/or solid adenocarcinoma. A panel of 15 radiologists was divided into 2 groups (group A, 9 radiologists; and group B, 6 radiologists). The 2 groups independently measured the maximal and perpendicular dimensions of the solid components and entire tumors on the lung and mediastinal window settings. The solid proportion of nodule was calculated by dividing the solid portion size (lung and mediastinal window settings) by the nodule size (lung window setting). The maximal dimensions of the invasive focus were measured on the corresponding pathologic specimens by 2 pathologists.

RESULTS

The solid proportion was larger in the following descending order: IVA-micropapillary and/or solid, IVA-acinar and/or papillary, IVA-lepidic, MIA, and AIS. For both groups A and B, a solid portion > 0.8 cm in the lung window setting or > 0.6 cm in the mediastinal window setting on CT was a significant indicator of pathologic invasiveness > 0.5 cm (P < .001; receiver operating characteristic analysis using Youden's index).

CONCLUSION

A solid portion > 0.8 cm on the lung window setting or solid portion > 0.6 cm on the mediastinal window setting on CT predicts for histopathologic invasiveness to differentiate IVA from MIA and AIS.