Three-Dimensional Ground Glass Opacity Ratio in CT Images Can Predict Tumor Invasiveness of Stage IA Lung Cancer

Artificial-intelligence-based computed tomography histogram analysis predicting tumor invasiveness of lung adenocarcinomas manifesting as radiological part-solid nodules

Background

Tumor invasiveness plays a key role in determining surgical strategy and patient prognosis in clinical practice. The study aimed to explore artificial-intelligence-based computed tomography (CT) histogram indicators significantly related to the invasion status of lung adenocarcinoma appearing as part-solid nodules (PSNs), and to construct radiomics models for prediction of tumor invasiveness.

Methods

We identified surgically resected lung adenocarcinomas manifesting as PSNs in Peking University People's Hospital from January 2014 to October 2019. Tumors were categorized as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC) by comprehensive pathological assessment. The whole cohort was randomly assigned into a training (70%, n=832) and a validation cohort (30%, n=356) to establish and validate the prediction model. An artificial-intelligence-based algorithm (InferRead CT Lung) was applied to extract CT histogram parameters for each pulmonary nodule. For feature selection, multivariate regression models were built to identify factors associated with tumor invasiveness. Logistic regression classifier was used for radiomics model building. The predictive performance of the model was then evaluated by ROC and calibration curves.

Results

In total, 299 AIS/MIAs and 889 IACs were included. In the training cohort, multivariate logistic regression analysis demonstrated that age [odds ratio (OR), 1.020; 95% CI, 1.004-1.037; p=0.017], smoking history (OR, 1.846; 95% CI, 1.058-3.221; p=0.031), solid mean density (OR, 1.014; 95% CI, 1.004-1.024; p=0.008], solid volume (OR, 5.858; 95% CI, 1.259-27.247; p = 0.037), pleural retraction sign (OR, 3.179; 95% CI, 1.057-9.559; p = 0.039), variance (OR, 0.570; 95% CI, 0.399-0.813; p=0.002), and entropy (OR, 4.606; 95% CI, 2.750-7.717; p<0.001) were independent predictors for IAC. The areas under the curve (AUCs) in the training and validation cohorts indicated a better discriminative ability of the histogram model (AUC=0.892) compared with the clinical model (AUC=0.852) and integrated model (AUC=0.886).

Conclusion

We developed an AI-based histogram model, which could reliably predict tumor invasiveness in lung adenocarcinoma manifesting as PSNs. This finding would provide promising value in guiding the precision management of PSNs in the daily practice.

Risk of adenocarcinoma in patients with a suspicious ground-glass opacity: a retrospective review

BACKGROUND

Both primary lung adenocarcinoma and benign processes can have a ground-glass opacity (GGO) appearance on imaging. This study evaluated the incidence of and risk factors for malignancy in a diverse cohort of patients who underwent resection of a GGO suspicious for lung cancer.

METHODS

All patients who underwent resection of a pulmonary nodule with a GGO component and suspected to be primary lung cancer at a single institution from 2001-2017 were retrospectively reviewed. Risk factors for malignancy were evaluated using multivariable logistic regression analysis that included nodule size, age, sex, and race as potential predictors.

RESULTS

The incidence of pulmonary adenocarcinoma in the 243 patients who met inclusion criteria was 86% (n=208). The most common pathologic findings in 35 patients with a benign pathology was granulomatous inflammation (n=14, 40%). Risk factors for adenocarcinoma in multivariable logistic regression were age [odds ratio (OR) 1.06, P=0.003], GGO size (OR 2.76, P<0.001), female sex (OR 4.47, P=0.002), and Asian race (OR 8.35, P=0.002). In this cohort, adenocarcinoma was found in 100% (44/44) of Asian females, 86% (25/29) of Asian males, 84% (98/117) of non-Asian females, and 77% (41/53) of non-Asian males.

CONCLUSIONS

The likelihood of adenocarcinoma in lung nodules with a ground-glass component is influenced by sex and race. Asian females with a GGO have a much higher likelihood of having adenocarcinoma than men and non-Asians. This data can be used when deciding whether to pursue nodule resection or surveillance in a patient with a GGO.

[Comparison of Two-dimensional and Three-dimensional Features of Chest CT in the Diagnosis of Invasion of Pulmonary Ground Glass Nodules]

BACKGROUND

At present, more and more studies predict invasive adenocarcinoma (IAC) through three-dimensional features of pulmonary nodules, but few studies have confirmed that three-dimensional features have more advantages in diagnosing IAC than traditional two-dimensional features of pulmonary nodules. This study analyzed the differences of chest computed tomography (CT) features between IAC and minimally invasive adenocarcinoma (MIA) from three-dimensional and two-dimensional levels, and compared the ability of diagnosing IAC. The non-invasive adenocarcinoma group includes precursor glandular lesions (PGL) and minimally invasive adenocarcinoma (MIA).

METHODS

The clinical data of 1,045 patients with ground glass opacity (GGO) from January to December 2019 were collected. Then the correlation between preoperative CT image characteristics and pathological results were analyzed retrospectively. The independent influencing factors for the identification of IAC were screened out according to two-dimensional and three-dimensional classification by multivariate Logistic regression and the cut-off point for the identification of IAC was found out through the receiver operating characteristic (ROC) curve. At last, the ability of diagnosing IAC was evaluated by Yoden index.

RESULTS

The diameter of nodule, the diameter of solid component, the diameter of mediastinal window nodule in two-dimensional factors, and the volume of nodule, the volume of solid part and the average CT value in three-dimensional factors were independent risk factors for the diagnosis of IAC. These factors were arranged by Yoden index: solid partial volume (0.601)>nodule volume (0.536)>solid component diameter (0.525)>nodule diameter (0.518)>mediastinal window nodule diameter (0.488)>proportion of solid component volume (0.471)>1-tumor disappearance ratio (TDR) (0.468)>consolidation tumor ratio (CTR) (0.394)>average CT value (0.380).

CONCLUSIONS

The CT features of three-dimensional are better than two-dimensional in the diagnosis of IAC, and the size of solid components is better than the overall size of nodules.

Early lung carcinogenesis and tumor microenvironment observed by single-cell transcriptome analysis

•

Tregs lead immune-evasive TME of early lung cancer of never smoker.

•

Depletion of γδT and NK cells and infiltration of B cells begins in early TME.

•

Early lung cancer cells were characterized by dysregulated surfactant pathway.

•

CAFs show enrichment of gene sets that inhibit vascular formation.

In tumor tissue, tip-like endothelial cells begin to be replaced with immature ones.

With the increasing interest in health screening with chest CT Ground-glass nodule (GGN) has become one of the common lung lesions encountered in daily medical practice. Because lung adenocarcinoma in the form of GGN is an ideal model for studying early lung carcinogenesis, 11 GGN and normal lung specimens from 6 never smoker patients were studied by single-cell RNA sequencing. Lung cancer cells showed enrichment of gene sets related to small vesicle processing and surfactant homeostasis compared to non-malignant lung epithelial cells, suggesting the dysregulation of surfactant pathway may be involved in early lung carcinogenesis. Along with cancer-associated fibroblasts showing enrichment of gene sets involved in negative regulation of protein kinase activity and negative regulation of endothelial cell proliferation, tumor microenvironment (TME) was dominated by infiltration of TNFRSF4+/TNFRSF18+/CTLA4+ regulatory T cells (Treg) and depletion of CD8+ cytotoxic T cells (TC) and γδTC. Majority of mucosa-associated lymphoid tissue B cells (BCs) and follicular BCs were detected within tumor tissue, which was associated with CXCL13 overexpressed in intratumoral Tregs and CD4+ memory TCs. Coordination of components of the TME towards immune evasion is governed by Tregs from the onset of lung cancer, requiring unremitting efforts to target and overcome them. This provision of information on changes in cancer cell-specific biomarkers and TME using early lung cancer from never smokers will provide new insight into early lung carcinogenesis and useful targets for treatment.

Radiomics nomogram: A noninvasive tool for preoperative evaluation of the invasiveness of pulmonary adenocarcinomas manifesting as ground-glass nodules

In this study, we aimed to establish a radiomics nomogram that noninvasively evaluates the invasiveness of pulmonary adenocarcinomas manifesting as ground-glass nodules (GGNs). Computed tomography (CT) images of 509 patients manifesting as GGNs were collected: 70% of cases were included in the training cohort and 30% in the validation cohort. The Max-Relevance and Min-Redundancy (mRMR) and the least absolute shrinkage and selection operator (LASSO) algorithm were used to select the radiomics features and construct a radiomics signature. Univariate and multivariate logistic regression were used to select the invasiveness-related clinical and CT morphological predictors. Age, smoking history, long diameter, and average CT value were retained as independent predictors of GGN invasiveness. A radiomics nomogram was established by integrating clinical and CT morphological features with the radiomics signature. The radiomics nomogram showed good predictive ability in the training set (area under the curve [AUC], 0.940; 95% confidence interval [CI], 0.916-0.964) and validation set (AUC, 0.946; 95% CI, 0.907-0.986). This radiomics nomogram may serve as a noninvasive and accurate predictive tool to determine the invasiveness of GGNs prior to surgery and assist clinicians in creating personalized treatment strategies.

Stepwise flowchart for decision making on sublobar resection through the estimation of spread through air space in early stage lung cancer1

OBJECTIVES

The sensitivity for tumor spread through air space (STAS), an independent risk factor for locoregional recurrence after sublobar resection for lung cancer, has been relatively low in frozen sections. We aimed to determine predictors with high negative predictive value for the presence of STAS and to provide the flowchart in combination with these predictors for the decision-making for sublobar resection.

MATERIALS AND METHODS

Between July 2015 and December 2017, 387 patients who underwent surgery for non-small cell lung cancer (NSCLC) with pathologic findings of the total masses measuring ≤ 2 cm were enrolled. The lesions were divided into two groups according to presence of STAS. We compared the preoperative characteristics, operative data, and developed a flowchart for STAS prediction using receiver operator characteristic curve analysis and multivariable logistic regression.

RESULTS

The STAS-positive group (N = 111) had a significantly higher preoperative tumor size (1.70 [1.5] vs 1.50 [0.69], p < 0.001) and standardized uptake value tumor-to-liver (SUV T/L) ratio (1.40 [1.60] vs 0.60 [1.10], p < 0.001) and a significantly lower two-dimensional ground-glass opacity (GGO) percentage (35.86 [61.00] vs 78.14 [39.00], p < 0.001). Meanwhile, the STAS-negative group (N = 286) had higher lepidic predominance (41.6% vs. 1.8%, p < 0.001). We developed a flowchart for predicting STAS in combination with two-dimensional GGO percentage on computed tomography (CT), SUV T/L ratio on positron-emission CT, and lepidic predominant pattern. The sensitivity, specificity, and negative predictive value for STAS positivity were 79.3%, 68.5%, and 89.5%, respectively.

CONCLUSIONS

The stepwise flowchart using two-dimensional GGO percentage on CT, maximum SUV, and lepidic predominance might be helpful in selecting patients with early NSCLC for sublobar resection.

Correlation Study of 18F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography in Pathological Subtypes of Invasive Lung Adenocarcinoma and Prognosis

Purpose: To investigate the correlation between ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) metabolic parameters and clinicopathological factors in pathological subtypes of invasive lung adenocarcinoma and prognosis.

Patients and Methods: Metabolic parameters and clinicopathological factors from 176 consecutive patients with invasive lung adenocarcinoma between August 2008 and August 2016 who underwent ¹⁸F-FDG PET/CT examination were retrospectively analyzed. Invasive lung adenocarcinoma was divided into five pathological subtypes:lepidic predominant adenocarcinoma (LPA), acinar predominant adenocarcinoma (APA), papillary predominant adenocarcinoma (PPA), solid predominant adenocarcinoma (SPA), and micropapillary predominant adenocarcinoma (MPA). The differences in metabolic parameters [maximal standard uptake value (SUVmax), mean standard uptake value (SUVmean), total lesion glycolysis (TLG), and metabolic tumor volume (MTV)] and tumor diameter for different pathological subtypes were analyzed. Patients were divided into two groups according to their prognosis: good prognosis group (LPA, APA, PPA) and poor prognosis group (SPA, MPA). Logistic regression was used to filter predictors and construct a predictive model, and areas under the receiver operating curve (AUC) were calculated. Cox regression analysis was performed on prognostic factors.

Results: 82 (46.6%) females and 94 (53.4%) males of patients with invasive lung adenocarcinoma were enrolled in this study. Metabolic parameters and tumor diameter of different pathological subtype had statistically significant (P < 0.05). The predictive model constructed using independent predictors (Distant metastasis, Ki-67, and SUVmax) had good classification performance for both groups. The AUC for SUVmax was 0.694 and combined with clinicopathological factors were 0.745. Cox regression analysis revealed that Stage, TTF-1, MTV, and pathological subtype were independent risk factors for patient prognosis. The hazard ratio (HR) of the poor prognosis group was 1.948 (95% CI 1.042–3.641) times the good prognosis group. The mean survival times of good and poor prognosis group were 50.2621 (95% CI 47.818–52.706) and 35.8214 (95% CI 27.483–44.159) months, respectively, while the median survival time was 47.00 (95% CI 45.000–50.000) and 31.50 (95% CI 23.000–49.000) months, respectively.

Conclusion: PET/CT metabolic parameters combined with clinicopathological factors had good classification performance for the different pathological subtypes, which may provide a reference for treatment strategies and prognosis evaluation of patients.

Application of artificial intelligence in the diagnosis of multiple primary lung cancer

Artificial intelligence (AI) based on deep learning, convolutional neural networks and big data has been increasingly effective in the diagnosis and treatment of multiple primary pulmonary nodules. In comparison to previous imaging systems, AI measures more objective parameters such as three‐dimensional (3D) volume, probability of malignant nodules, and possible pathological patterns, making the access to the properties of nodules more objective. In our retrospective study, a total of 53 patients with synchronous and metachronous multiple pulmonary nodules were enrolled of which 33 patients were confirmed by pathological tests to have primary binodules, and nine to have primary trinodules. A total of 15 patients had only one focus removed. The statistical results showed that the agreement in the AI diagnosis and postoperative pathological tests was 88.8% in identifying benign or malignant lesions. In addition, the probability of malignancy of benign lesions, preinvasive lesions (AAH, AIS) and invasive lesions (MIA, IA) was totally different (49.40±38.41% vs 80.22±13.55% vs 88.17±17.31%). The purpose of our study was to provide references for the future application of AI in the diagnosis and follow‐up of multiple pulmonary nodules. AI may represent a relevant diagnostic aid that shows more accurate and objective results in the diagnosis of multiple pulmonary nodules, reducing the time required for interpretation of results by directly displaying visual information to doctors and patients and together with the clinical conditions of MPLC patients, offering plans for follow‐up and treatment that may be more beneficial and reasonable for patients. Despite the great application potential in pneumosurgery, further research is needed to verify the accuracy and range of the application of AI.

Changes in quantitative parameters of pulmonary nonsolid nodule induced by lung inflation according to paired inspiratory and expiratory computed tomography imaging

OBJECTIVE

To evaluate quantitative parameters of nonsolid nodules on paired inspiratory and expiratory computed tomography (CT) and to examine whether these parameters are sensitive to lung inflation reflected by lung volume.

METHODS

Thirty-three patients with 41 nonsolid nodules were included in this prospective study. Paired inspiratory and low-dose respiratory plain chest CT were performed. The volume and density of nonsolid nodule(s), both lungs, the right and left lung, and five lobes, were analyzed in inspiratory and expiratory CT scans. The ratio of expiratory to inspiratory parameters was calculated and labeled as parameter_(E-I)/I. To standardize the changes in nonsolid nodule quantitative parameters, the ratio of nonsolid nodule parameter to lung parameter was also calculated. Quantitative parameters were compared between inspiratory and expiratory CT.

RESULTS

Nonsolid nodule volumes on expiratory CT were reduced by 19.8% ± 12.9%, while the density was increased by 11.4% ± 8.8%. The volume of nonsolid nodules was significantly greater on inspiratory compared with expiratory CT (p < 0.001). The density of nonsolid nodules was significantly greater on expiratory than inspiratory CT (p < 0.001). The volume_(E-I)/I was significantly greater than density_(E-I)/I both in nonsolid nodules and lung. The volume_(E-I)/I and density_(E-I)/I of nonsolid nodules were independent of size. The density_(E-I)/I of nonsolid nodule was greater in the lower lobe than that in the upper lobe (p = 0.002).

CONCLUSION

Volume changes in nonsolid nodules were more sensitive than density changes in expiratory phase. The density of lower lobe nodules was more susceptible to respiration. Expiratory scanning is not recommended for quantification of nonsolid nodules and/or follow-up.

KEY POINTS

• The nonsolid nodule volume on expiratory CT was reduced by 19.8% ± 12.9%. • The nonsolid nodule density on expiratory CT was increased by 11.4% ± 8.8%. • The volume _(E-I)/I and density _(E-I)/I of nonsolid nodules were independent of size.

Radiomics signature: a biomarker for the preoperative discrimination of lung invasive adenocarcinoma manifesting as a ground-glass nodule

OBJECTIVES

To identify the radiomics signature allowing preoperative discrimination of lung invasive adenocarcinomas from non-invasive lesions manifesting as ground-glass nodules.

METHODS

This retrospective primary cohort study included 160 pathologically confirmed lung adenocarcinomas. Radiomics features were extracted from preoperative non-contrast CT images to build a radiomics signature. The predictive performance and calibration of the radiomics signature were evaluated using intra-cross (n=76), external non-contrast-enhanced CT (n=75) and contrast-enhanced CT (n=84) validation cohorts. The performance of radiomics signature and CT morphological and quantitative indices were compared.

RESULTS

355 three-dimensional radiomics features were extracted, and two features were identified as the best discriminators to build a radiomics signature. The radiomics signature showed a good ability to discriminate between invasive adenocarcinomas and non-invasive lesions with an accuracy of 86.3%, 90.8%, 84.0% and 88.1%, respectively, in the primary and validation cohorts. It remained an independent predictor after adjusting for traditional preoperative factors (odds ratio 1.87, p < 0.001) and demonstrated good calibration in all cohorts. It was a better independent predictor than CT morphology or mean CT value.

CONCLUSIONS

The radiomics signature showed good predictive performance in discriminating between invasive adenocarcinomas and non-invasive lesions. Being a non-invasive biomarker, it could assist in determining therapeutic strategies for lung adenocarcinoma.

KEY POINTS

• The radiomics signature was a non-invasive biomarker of lung invasive adenocarcinoma. • The radiomics signature outweighed CT morphological and quantitative indices. • A three-centre study showed that radiomics signature had good predictive performance.

Solitary ground-glass opacity nodules of stage IA pulmonary adenocarcinoma: combination of 18F-FDG PET/CT and high-resolution computed tomography features to predict invasive adenocarcinoma

To investigate the performance of combined 18F-FDG Positron Emission Tomography/Computed Tomography with high-resolution CT for differentiating invasive adenocarcinoma from adenocarcinoma in situ (pre-invasive lesion) or minimally invasive adenocarcinoma in stage IA lung cancer patients with solitary ground-glass opacity nodules. This retrospective study enrolled 58 consecutive stage IA pulmonary adenocarcinoma patients with solitary ground-glass opacity nodules. The characteristics and measurements of the ground-glass opacity nodules as pure ground-glass opacity nodules and mixed ground-glass opacity nodules in the pre-invasive or minimally invasive adenocarcinoma and invasive adenocarcinoma groups on Positron Emission Tomography/Computed Tomography and high-resolution CT were compared and analyzed. Ground-glass opacity nodules in the pre-invasive or minimally invasive adenocarcinoma group preferentially manifested as pure ground-glass opacity nodule (p < 0.01) compared to the invasive adenocarcinoma group. While cystic appearance was more common in the invasive adenocarcinoma group (p < 0.05). Significant differences were found in the diameter of the ground-glass opacity nodule itself and its solid component, and consolidation/tumor ratio between the two groups. The sensitivity in predicting invasive adenocarcinoma was higher with a combined consolidation/tumor ratio > 0.38 and SUV_max > 1.46 in mixed ground-glass opacity nodule when compared to those of SUV_max > 0.95 alone or consolidation/tumor ratio> 0.39 alone (both p > 0.05). For a mixed ground-glass opacity nodule combined consolidation/tumor ratio > 0.38 and SUV_max > 1.46 appears to better predict invasive adenocarcinoma in stage IA lung cancer patients with solitary ground-glass opacity nodules.