Unraveling the Significance of MET Focal Amplification in Lung Cancer: Integrative NGS, FISH, and IHC Investigation.

MET amplification (METamp) represents a promising therapeutic target in non-small cell lung cancer, but no consensus has been established to identify METamp-dependent tumors that could potentially benefit from MET inhibitors. In this study, an analysis of MET amplification/overexpression status was performed in a retrospectively recruited cohort comprising 231 patients with non-small cell lung cancer from Shanghai Chest Hospital (SCH cohort) using 3 methods: fluorescence in situ hybridization (FISH), hybrid capture-based next-generation sequencing, and immunohistochemistry for c-MET and phospho-MET. The SCH cohort included 130 cases known to be METamp positive by FISH and 101 negative controls. The clinical relevance of these approaches in predicting the efficacy of MET inhibitors was evaluated. Additionally, next-generation sequencing data from another 2 cohorts including 22,010 lung cancer cases were utilized to examine the biological characteristics of different METamp subtypes. Of the 231 cases, 145 showed MET amplification/overexpression using at least 1 method, whereas only half of them could be identified by all 3 methods. METamp can occur as focal amplification or polysomy. Our study revealed that the inconsistency between next-generation sequencing and FISH primarily occurred in the polysomy subtype. Further investigations indicated that compared with polysomy, focal amplification correlated with fewer co-occurring driver mutations, higher protein expressions of c-MET and phospho-MET, and higher incidence in acquired resistance than in de novo setting. Moreover, patients with focal amplification presented a more robust response to MET inhibitors compared with those with polysomy. Notably, a strong correlation was observed between focal amplification and programmed cell death ligand-1 expression, indicating potential therapeutic implications with combined MET inhibitor and immunotherapy for patients with both alterations. Our findings provide insights into the molecular complexity and clinical relevance of METamp in lung cancer, highlighting the role of MET focal amplification as an oncogenic driver and its feasibility as a primary biomarker to further investigate the clinical activity of MET inhibitors in future studies.

Distinct mutational features across preinvasive and invasive subtypes identified through comprehensive profiling of surgically resected lung adenocarcinoma.

Lung adenocarcinoma (LUAD) is a heterogeneous disease. Our study aimed to understand the unique molecular features of preinvasive to invasive LUAD subtypes. We retrospectively analyzed the clinical, histopathological, and molecular data of 3,254 Chinese patients with preinvasive lesions (n = 252), minimally invasive adenocarcinomas (n = 479), and invasive LUAD (n = 2,523). Molecular data were elucidated using a targeted 68-gene next-generation sequencing panel. Our findings revealed four preinvasive lesion-predominant gene mutations, including MAP2K1 insertion-deletions (indels), BRAF non-V600E kinase mutations, and exon 20 insertions (20ins) in both EGFR and ERBB2, which we referred to as mutations enriched in AIS (MEA). The detection rate of MEA in invasive tumors was relatively lower. MAP2K1 missense mutations, which were likely passenger mutations, co-occurred with oncogenic driver mutations, while small indels were mutually exclusive from other genes regardless of the invasion level. BRAF non-V600E kinase-mutant invasive adenocarcinomas (IAC) had significantly higher mutation rates in tumor suppressor genes but lower frequency of co-occurring oncogenic driver mutations than non-kinase-mutant IAC, suggesting the potential oncogenic activity of BRAF non-V600E kinase mutations albeit weaker than BRAF V600E. Moreover, similar to the extremely low frequency of MAP2K1 indels in IAC, BRAF non-V600E kinase domain mutations co-occurring with TSC1 mutations were exclusively found in preinvasive lesions. Compared with EGFR L858R and exon 19 deletion, patients with preinvasive lesions harboring 20ins in either EGFR or ERBB2 were significantly younger, while those with IAC had similar age. Furthermore, our study demonstrated distinct mutational features for subtypes of oncogene mutations favored by different invasion patterns in adenocarcinomas. In conclusion, our data demonstrate distinct mutational features between preinvasive lesions and invasive tumors with MEA, suggesting the involvement of MEA in the early stages of tumorigenesis. Further pre-clinical studies are required to establish the role of these genes in the malignant transformation of LUAD.

The possibility of endoscopic treatment of cN0 submucosal esophageal cancer: results from a surgical cohort.

BACKGROUND: We analyzed the pathological characteristics and recurrence pattern of cN0 submucosal esophageal cancer after esophagectomy and conducted risk stratification to determine the feasibility of performing endoscopic resection for cN0pT1b esophageal squamous cell malignancies. METHODS: We retrospectively enrolled 167 patients who underwent right-sided transthoracic esophagectomy and extended thoracic/abdominal two-field lymphadenectomy. Patients with pathologically confirmed lymph node metastasis or tumor recurrence constituted the high-risk group for endoscopic submucosal resection, and the remainder were defined as low risk. Factors affecting lymphatic metastasis and long-term recurrence were identified by univariate and multivariate analyses. RESULTS: Postoperative pathology showed that five patients (5/167; 3%) had lymph node metastases. Follow-up ranged from 12-60 months, with a median of 29 months. A total of 17 patients (10.2%) had recurrences during follow-up, including three patients with pathologic nodal metastasis (pN +) found at surgery. Invasion depth, differentiation, and tumor size differed significantly in high-risk patients. Overall 3-year survival rates were 94.2% (low-risk) and 40.9% (high-risk) (p < 0.01). Twenty-one patients with sm1 cancer, high tumor differentiation, and tumor length < 2 cm had no lymph node metastasis or lymphovascular invasion, and none of these patients experienced recurrence. CONCLUSIONS: Endoscopic submucosal resection alone may be feasible for patients with small (≤ 2 cm) clinically N0 submucosal esophageal squamous cell carcinoma with low invasion depth (sm1) and higher differentiation, but prospective studies are required for confirmation. Other patients require surgical resection with extended two-field thoracic/abdominal lymphadenectomy.

Near-Infrared Fluorescent Image-Guided Lymphatic Mapping in Esophageal Squamous Cell Carcinoma.

BACKGROUND: Recently, the feasibility of near-infrared (NIR) image-guided sentinel lymph node (SLN) mapping has been tested in patients with gastrointestinal cancer. The aim of this study is to investigate whether SLN mapping can be used to identify mediastinal lymph node metastases during minimally invasive esophagectomy and explore the lymphatic drainage pattern of esophageal squamous cell carcinoma (ESCC) using NIR fluorescent imaging. PATIENTS AND METHODS: A total of 21 patients diagnosed with cT1-3 stage ESCC were enrolled. Patients received submucosal injection of indocyanine green diluted with sodium chloride (0.9%) at the start of the esophagectomy procedure followed by NIR mapping. RESULTS: Thoracoscopic-assisted McKeown esophagectomy with NIR imaging was successfully performed in all patients. The detection rate and number of NIR+ lymph nodes were 95.2% (20/21) and 4.0 (2.0-6.5), respectively. The accuracy, false-negative rates, and negative predictive value were 100% (10 of 10 cases), 0% (0 of 4), and 100% (6 of 6), respectively, for pT1/T2 diseases; and 80.0% (8 of 10), 40% (2 of 5), and 71.4% (5 of 7), respectively, for pT3 diseases. The NIR+ region was the most commonly detected in the right recurrent laryngeal nerve (80%), and the NIR+ region was identified in the upper mediastinal zone in 20 patients. CONCLUSIONS: Evaluation of the lymphatic drainage pattern and the application of sentinel lymph node in ESCC with real-time NIR imaging could be effective, especially in pT1/2 disease. NIR imaging-guided SLN navigation appears to be a clinically beneficial less-invasive method for treating ESCC.

Malignant perivascular epithelioid cell tumor of the lung synchronous with a primary adenocarcinoma: one case report and review of the literature.

BACKGROUND: Perivascular Epithelioid Cell Tumors (PEComa) is an extraordinarily rare mesenchymal neoplasm especially the malignant type originating from the lung. To date, only 8 cases of malignant or malignant potential pulmonary PEComa had been documented. Firm diagnostic criteria for malignant pulmonary PEComa need urgently to be established. CASE PRESENTATION: We report a challenging case of malignant pulmonary PEComa combined with a primary adenocarcinoma in a 54-year-old man. The PEComa-like tumor showed strong Melan-A and weak transcription factor E3 (TFE3) protein expression but no TFE3 gene rearrangement. The carcinoma-like nodule was recognized as a poorly differentiated primary lung adenocarcinoma. DISCUSSION AND CONCLUSIONS: Our case report was the first case of malignant pulmonary PEComa synchronous with a primary adenocarcinoma and studied the dilemma of diagnosing benign versus malignant criteria for this uncommon tumor.

[Detection of ROS1 fusion gene in pulmonary adenocarcinoma and its clinicopathologic features].

OBJECTIVE: To detect the presence of ROS1 fusion gene in pulmonary adenocarcinoma and its clinicopathologic parameters. METHODS: Fluorescence RT-PCR was used to detect the presence of ROS1 fusion gene in 369 surgical resection samples of pulmonary adenocarcinoma with known EGFR mutation status. The presence of ROS1 fusion gene in correlation with clinicopathologic features was analyzed. Sixteen positive and 20 negative samples by RT-PCR were further confirmed by direct sequencing. RESULTS: ROS1 fusion gene was detected in 16 of 369 lung adenocarcinoma samples (4.3%). The presence of ROS1 fusion gene was not correlated to gender, age, smoking history, tumor site, size, histological subtype, tumor differentiation, T staging, lymph node metastasis, TNM staging and EGFR mutation (P > 0.05). The frequency of ROS1 fusion gene was similar in female and male patients, 4.4% (8/183) vs 4.3% (8/186), P > 0.05. The presence of ROS1 fusion gene in patients of ≤ 60 years of age was higher than that in patients of > 60 years, 5.1% (10/195) vs 3.4% (6/174), P > 0.05. The rate of ROS1 fusion gene of non-smokers was a slight higher than that of smokers, 4.4% (14/318) vs 3.9% (2/51), P > 0.05. Both positive and negative cases were confirmed by direct sequencing in all cases. CONCLUSIONS: ROS1 fusion gene occurs more frequently in younger and non-smoking patients of pulmonary adenocarcinoma, and may coexist with EGFR mutations. ROS1 fusion gene seems to define a distinct subset of pulmonary adenocarcinoma.

Proteomic characterization of esophageal squamous cell carcinoma response to immunotherapy reveals potential therapeutic strategy and predictive biomarkers.

Immunotherapy is the first-line therapy for esophageal squamous cell carcinoma (ESCC), yet many patients do not respond due to drug resistance and the lack of reliable predictive markers. We collected 73 ESCC patients (including discovery cohort and validation cohort) without immune thrombocytopenia and undergoing anti-PD1 immunotherapy. Proteomic and phosphoproteomic analysis of 73 ESCC treatment-naive samples by mass spectrometry-based label-free quantification were applied to explore the potential resistant and sensitive mechanisms, and identify predictive markers of ESCC immunotherapy. Comparative analysis found the pathways related to immune and mitochondrial functions were associated with ESCC immunotherapy sensitivity; while platelet activation bioprocess showed negative correlation with CD8+ T cells and related to ESCC immunotherapy non-sensitivity. Finally, we identified 10 ESCC immunotherapy predictive biomarkers with high accuracy (≥ 0.90) to predict the immunotherapeutic response, which was validated in the independent cohort.

Molecular, clinicopathological characteristics and surgical results of resectable SMARCA4-deficient thoracic tumors.

PURPOSE: SMARCA4-deficient thoracic tumors are rapid aggressive malignancies, often diagnosed at an advanced and inoperable stage. The value of pulmonary resection for resectable SMARCA4-deficient thoracic tumors is largely unknown. METHODS: In this observational study, we included 45 patients who received surgery for stage I-III SMARCA4-deficient tumors. We compared the molecular, clinicopathological characteristics and survival between SMARCA4-dNSCLC and SMARCA4-deficient undifferentiated tumor (SMARCA4-dUT) patients. RESULTS: Thirty-four SMARCA4-dNSCLC and 11 SMARCA4-dUT patients were included in this study. Molecular profiles were available in 33 out of 45 patients. The most common mutated gene was TP53 (21, 64%), and followed by STK11 (9, 27%), KRAS (5, 15%), FGFR1 (4, 12%) and ROS1 (4, 12%). There were 3 patients that harbored ALK mutation including 1 EML4-ALK rearrangement. There were 2 patients that harbored EGFR rare site missense mutation. SMARCA4-dUT patients had significance worse TTP (HR = 4.35 95% CI 1.77-10.71, p = 0.001) and OS (HR = 4.27, 95% CI 1.12-16.35, p = 0.022) compared to SMARCA4-dNSCLC patients. SMARCA4-dUT histologic type, stage II/III, R1/2 resection and lymphovascular invasion were independent poor prognostic predictors for both TTP and OS. There were 8 patients who received immunotherapy, the objective response rate was 50%. The SMARCA4-dNSCLC patient with ALK rearrangement was treated with crizotinib as second-line therapy, and achieved stable disease for 9.7 months. CONCLUSION: Patients with SMARCA4-deficient tumors have a high probability of early recurrence after surgery, except for stage I patients. Immunotherapy seems to be a valuable strategy to treat recurrence.

CoADS: Cross attention based dual-space graph network for survival prediction of lung cancer using whole slide images.

BACKGROUND AND OBJECTIVE: Accurate overall survival (OS) prediction for lung cancer patients is of great significance, which can help classify patients into different risk groups to benefit from personalized treatment. Histopathology slides are considered the gold standard for cancer diagnosis and prognosis, and many algorithms have been proposed to predict the OS risk. Most methods rely on selecting key patches or morphological phenotypes from whole slide images (WSIs). However, OS prediction using the existing methods exhibits limited accuracy and remains challenging. METHODS: In this paper, we propose a novel cross-attention-based dual-space graph convolutional neural network model (CoADS). To facilitate the improvement of survival prediction, we fully take into account the heterogeneity of tumor sections from different perspectives. CoADS utilizes the information from both physical and latent spaces. With the guidance of cross-attention, both the spatial proximity in physical space and the feature similarity in latent space between different patches from WSIs are integrated effectively. RESULTS: We evaluated our approach on two large lung cancer datasets of 1044 patients. The extensive experimental results demonstrated that the proposed model outperforms state-of-the-art methods with the highest concordance index. CONCLUSIONS: The qualitative and quantitative results show that the proposed method is more powerful for identifying the pathology features associated with prognosis. Furthermore, the proposed framework can be extended to other pathological images for predicting OS or other prognosis indicators, and thus delivering individualized treatment.

Hierarchical multimodal fusion framework based on noisy label learning and attention mechanism for cancer classification with pathology and genomic features.

Classification of subtype and grade is imperative in the clinical diagnosis and prognosis of cancer. Many deep learning-based studies related to cancer classification are based on pathology and genomics. However, most of them are late fusion-based and require full supervision in pathology image analysis. To address these problems, we present an integrated framework for cancer classification with pathology and genomics data. This framework consists of two major parts, a weakly supervised model for extracting patch features from whole slide images (WSIs), and a hierarchical multimodal fusion model. The weakly supervised model can make full use of WSI labels, and mitigate the effects of label noises by the self-training strategy. The generic multimodal fusion model is capable of capturing deep interaction information through multi-level attention mechanisms and controlling the expressiveness of each modal representation. We validate our approach on glioma and lung cancer datasets from The Cancer Genome Atlas (TCGA). The results demonstrate that the proposed method achieves superior performance compared to state-of-the-art methods, with the competitive AUC of 0.872 and 0.977 on these two datasets respectively. This paper establishes insight on how to build deep networks on multimodal biomedical data and proposes a more general framework for pathology image analysis without pixel-level annotation.

Single-cell profiling reveals the trajectory of FOLR2-expressing tumor-associated macrophages to regulatory T cells in the progression of lung adenocarcinoma.

An immunosuppressive microenvironment enriched with regulatory CD4+ T lymphocytes (Tregs) facilitates the progression of lung adenocarcinoma (LUAD). This study aims to investigate the cellular mechanism underlying the formation of the immunosuppressive microenvironment in LUAD. LUAD samples (n = 12) and normal lung samples (n = 3) were obtained from patients with different pathological stages of LUAD. Single-cell RNA sequencing was performed to classify cellular components and analyze the transcriptomes, including transcription factors/targets and chemokine ligands/receptors, followed by bioinformatics study such as pseudotime analysis. Myeloid cells and T cells were the most abundant cell types in tumors and normal lung tissues, while tumor-associated macrophage-folate receptor 2 (TAM-FOLR2) and CD4+ nuclear receptor subfamily 4 group A member 3 (NR4A3) exhibited sharp increases in invasive adenocarcinoma (IA). The enrichment of TAM-FOLR2 in IA might result from alveolar resident macrophage-resistin (ARM-RETN) transformation and recruitment of dendritic cells (DCs) and other TAMs, as evidenced by temporal trajectories and differential expression profiles of chemokine ligands/receptors versus those in the early stages of tumors. High expression of CCL17/19/22 was observed in IA as well as in DCs, along with the strong interaction of TAM-FOLR2 with DCs. The results of pseudotime analysis suggested that CD4+NR4A3 might potentially convert to CD4+FOXP3, further supported by the high expression of NR4A3 target genes in CD4+FOXP3 cells. This study provides a single-cell transcriptome atlas from preinvasive to invasive LUAD and reveals a potential ARM-RETN/TAM-FOLR2/DCs/CD4+NR4A3/CD4+FOXP3 trajectory in shaping the immune suppressive microenvironment along the pathogenesis of LUAD.

Identification and Validation of Noncanonical RET Fusions in Non-Small-Cell Lung Cancer through DNA and RNA Sequencing.

RET fusion has emerged as a targetable driver in non-small-cell lung cancer. A comparative analysis on RET fusions at DNA [DNA sequencing (DNA-seq)] and RNA [RNA sequencing (RNA-seq)] levels was performed in this study. Archived tumor samples from 54 non-small-cell lung cancer patients with DNA-level noncanonical RET fusions were selected for RNA-seq. RNA-seq identified RET fusion transcripts in 41 of 44 samples passing quality control. In the subset of cases harboring RET 3'-end fusions and predicted to produce in-frame proteins (group A; n = 33), RNA-seq identified the same 3'-end fusions in 32 (96.9%). A total of 26 of 32 also had a reciprocal RET 5'-end fusion detected by DNA-seq that was not transcribed. In the subset with DNA-level out-of-frame RET fusions (group B; n = 9), RNA-seq identified in-frame RET fusion transcripts in 8 cases (88.9%). In the subset only identified with a RET 5'-end fusion by DNA-seq (group C; n = 2), RNA-seq detected the corresponding 3'-end fusion in one case. The discordant DNA- and RNA-level fusions observed in group B may be mediated by complex genomic rearrangement events and transcriptional or post-transcriptional processes. In conclusion, DNA-seq demonstrates a high concordance of 96.9% on detecting in-frame RET fusion, but shows a low concordance on detecting out-of-frame RET fusion and RET 5'-end fusion compared with RNA-seq.

Decreased green autofluorescence of lung parenchyma is a biomarker for lung cancer tissues.

It is highly valuable to discover novel biomarkers for differentiating noninvasively the cancerous tissues from the nonneoplastic tissues of lung cancer. In current study, we determined the green autofluorescence (AF) of the pulmonary parenchyma of lung cancer patients, indicating that decreased green AF of pulmonary parenchyma may be the biomarker of this type: First, the green AF intensity of the cancerous tissues was significantly lower than that of the nonneoplastic tissues of the lung cancer patients; second, the green AF intensity of the nonneoplastic tissues of the lung squamous cell carcinoma was significantly lower than that of the lung adenocarcinoma; and third, "decreased green AF intensity" could be used for differentiating the nonneoplastic tissues and the cancerous tissues. Collectively, our study has suggested that decreased green AF of lung parenchyma is a biomarker for differentiating the cancerous tissues from the nonneoplastic tissues of lung cancer.

Challenging of frozen diagnoses of small sclerosing pneumocytoma.

AIMS: An increasing number of small pulmonary nodules are being screened by CT, and an intraoperative diagnosis is necessary for preventing excessive treatment. However, there is limited literature on the frozen diagnosis of small sclerosing pneumocytomas (SPs). In particular, tumours smaller than 1 cm are challenging for pathologists performing intraoperative frozen diagnosis. METHODS: In total, 230 cases of SP were surgically resected between January 2015 and March 2019 at Shanghai Chest Hospital, and of them, 76 cases were smaller than 1 cm. The histology and clinical information of these 76 cases (33.0%, 76/230) were reviewed retrospectively, 54 cases of which were diagnosed intraoperatively, and the pitfalls were summarised. All diagnoses were confirmed on permanent sections and immunohistochemical sections. RESULTS: Histologically, 78.9% (60/76) of the small SP was dominated by one growth pattern, and solid and papillary growth pattern were the most commonly misdiagnosed circumstances. The rate of intraoperative misdiagnosis of these SP smaller than 1 cm was 11.1% (6/54). CONCLUSIONS: The main reason for misdiagnosis was failure to recognise the dual cell populations and the cellular atypia. Diagnostic clues include the gross morphology, the presence of dual-cell populations and a hypercellular papillary core, foam cell accumulation in glandular spaces and haemorrhage and haemosiderin on the periphery. In spite of awareness of pitfalls some cases may still be essentially impossible to diagnose on frozen section.

Lung cancer subtype classification using histopathological images based on weakly supervised multi-instance learning.

Objective.Subtype classification plays a guiding role in the clinical diagnosis and treatment of non-small-cell lung cancer (NSCLC). However, due to the gigapixel of whole slide images (WSIs) and the absence of definitive morphological features, most automatic subtype classification methods for NSCLC require manually delineating the regions of interest (ROIs) on WSIs.Approach.In this paper, a weakly supervised framework is proposed for accurate subtype classification while freeing pathologists from pixel-level annotation. With respect to the characteristics of histopathological images, we design a two-stage structure with ROI localization and subtype classification. We first develop a method called multi-resolution expectation-maximization convolutional neural network (MR-EM-CNN) to locate ROIs for subsequent subtype classification. The EM algorithm is introduced to select the discriminative image patches for training a patch-wise network, with only WSI-wise labels available. A multi-resolution mechanism is designed for fine localization, similar to the coarse-to-fine process of manual pathological analysis. In the second stage, we build a novel hierarchical attention multi-scale network (HMS) for subtype classification. HMS can capture multi-scale features flexibly driven by the attention module and implement hierarchical features interaction.Results.Experimental results on the 1002-patient Cancer Genome Atlas dataset achieved an AUC of 0.9602 in the ROI localization and an AUC of 0.9671 for subtype classification.Significance.The proposed method shows superiority compared with other algorithms in the subtype classification of NSCLC. The proposed framework can also be extended to other classification tasks with WSIs.

EGFR Thr790Leu as a Potential Resistance Mechanism to First-Generation EGFR Tyrosine Kinase Inhibitor May Respond to Osimertinib in Patients With Lung Adenocarcinoma.

INTRODUCTION: It has been well established that EGFR Thr790Met is one of the major resistance mechanisms to first- and second-generation EGFR tyrosine kinase inhibitors (TKIs). Nevertheless, whether EGFR Thr790Leu (T790L), which shares the mutation site of Thr790 with EGFR Thr790Met, mediates resistance to EGFR TKIs remains elusive. The treatment options for patients harboring this rare mutation have not been reported. METHODS: Capture-based targeted ultradeep sequencing was performed on tumor and plasma samples collected at various treatment milestones from three patients with advanced lung adenocarcinoma undergoing targeted therapy. RESULTS: Needle biopsy of lymph node metastasis from patient 1 revealed EGFR T790L at disease progression on first-line treatment of gefitinib. Patient 2 had EGFR T790L identified from needle biopsy of lung tissue at disease progression on icotinib treatment. This patient was subsequently treated with osimertinib and achieved stable disease with a progression-free survival of 9 months. For patient 3, at disease recurrence after surgery, resected lung tumor tissue was retrieved for molecular profiling and revealed EGFR exon 19 deletion and EGFR T790L. The patient subsequently received osimertinib treatment and continued to benefit for 16 months and counting. She has maintained stable disease at the time of submission of this manuscript. CONCLUSIONS: We revealed for the first time that EGFR T790L may serve as a potential resistance mechanism to first-generation EGFR TKIs. We also report the first clinical evidence of efficacy generated by osimertinib in patients with lung adenocarcinoma harboring primary or acquired EGFR T790L, shedding light on treatment options for this subset of patients.

SUVmax of 18FDG PET/CT Predicts Histological Grade of Lung Adenocarcinoma.

OBJECTIVES: The relationship between the 18FDG PET-CT maximum standard uptake value (SUVmax) and the type of lung adenocarcinoma is still not established. The aim of this study was to investigate the relationship between SUVmax value and histological grade and pathological subtype of lung adenocarcinoma, and to determine the optimum SUVmax cutoffs for distinguishing different histological grades. MATERIALS AND METHODS: The data of 618 lung adenocarcinoma patients were retrospectively analyzed. The relationship between SUVmax measured on preoperative 18FDG-PET-CT and the histological grade and pathological subtype was examined. The Kruskal-Wallis test was used to compare differences among groups, and the Bonferroni-Dunn test for pairwise comparison among groups. ROC analysis was applied to determine the optimal cut-off values for distinguishing different groups. In addition, the cut-off value was verified in an independent cohort of 85 consecutive lung adenocarcinoma cases. RESULTS: The SUVmax was significantly different between the low, intermediate, and high-grade groups(p < .001). SUVmax value increased with increase in the degree of malignancy. The optimal cut-off value for identifying low-grade tumors was 2.01 (sensitivity 90.4%, specificity 86.9%, area under the curve [AUC] = 0.928, 95% confidence interval: 0.91-0.95; p < .001). The optimal cutoff SUVmax value for identifying high-grade tumors was 7.41 (sensitivity 79.8%, specificity 73.5%, AUC = 0.830, 95% confidence interval: 0.79-0.87; p < .001). The validation experiment showed that the coincidence rate was 88.89% in the low-level group, 64.15% in the middle-level group, and 78.57% in the high-level group. CONCLUSION: SUVmax can be used to predict pathological subtype and histological grade of lung adenocarcinoma. Thus, 18FDG PET-CT can serve as a noninvasive tool for precise diagnosis and help in the preoperative formulation of patient-specific treatment strategies.

A Weak Supervision-based Framework for Automatic Lung Cancer Classification on Whole Slide Image.

Classification of normal lung tissue, lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) by pathological images is significant for clinical diagnosis and treatment. Due to the large scale of pathological images and the absence of definitive morphological features between LUAD and LUSC, it is time-consuming, laborious and challenging for pathologists to analyze the microscopic histopathology slides by visual observation. In this paper, a pixel-level annotation-free framework was proposed to classify normal tissue, LUAD and LUSC slides. This framework can be divided into two stages: tumor classification and localization, and subtype classification. In the first stage, EM-CNN was utilized to distinguish tumor slides from normal tissue slides and locate the discriminative regions for subsequent analysis with only image-level labels provided. In the second stage, a multi-scale network was proposed to improve the accuracy of subtype classification. This method achieved an AUC of 0.9978 for tumor classification and an AUC of 0.9684 for subtype classification, showing its superiority in lung pathological image classification compared with other methods.

Molecular Profiling for Supernatants and Matched Cell Pellets of Pleural Effusions in Non-Small-Cell Lung Cancer.

Pleural effusion (PE) is commonly observed in advanced lung cancer patients. Cell-free total nucleic acid (cfTNA) isolated from cancer patients' plasma has allowed noninvasive tumor genome analyses; however, there are limited studies of detection and characterization of cfTNA in PE. Herein, we included 47 advanced non-small-cell lung cancer patients with PE, who had lung cancer driver mutations tested on tumor tissue specimens either at diagnosis or during disease progression. The supernatant and cell pellet of each PE were evaluated for molecular profiles in parallel on an Ion Torrent next-generation sequencing platform. Somatic mutations were detected in 89.1% supernatant cfTNA, but in only 54.3% of cell pellets. The overall concordance rate between supernatants and formalin-fixed, paraffin-embedded cell pellets at the mutation level was 53.3%. By contrast, 41.7% and 5.0% of somatic alterations were detected in supernatants and cell pellets, respectively. Furthermore, joint analysis of supernatants and cell pellets from PE showed a high concordance (88.3%) of variant detection with their respective tumor tissue specimens. Low-frequency T790M mutations in three cases (0.29%, 0.41%, and 1.56%) were detected in supernatants but not in the matched cell pellets or tumor tissues. In conclusion, pleural effusion-derived cfTNA can effectively be used in clinical practice for molecular analysis by next-generation sequencing, even in cases where corresponding cell pellets or tumor tissues yield insufficient material.