Non-small Cell Lung Cancer Datasets Research Articles

Effect of the STK11 mutation on therapeutic efficacy and prognosis in patients with non-small cell lung cancer: a comprehensive study based on meta-analyses and bioinformatics analyses

BackgroundThis study aimed to systematically analyze the effect of a serine/threonine kinase (STK11) mutation (STK11mut) on therapeutic efficacy and prognosis in patients with non-small cell lung cancer (NSCLC).MethodsCandidate articles were identified through a search of relevant literature published on or before April 1, 2023, in PubMed, Embase, Cochrane Library, CNKI and Wanfang databases. The extracted and analyzed data included the hazard ratios (HRs) of PFS and OS, the objective response rate (ORR) of immune checkpoint inhibitors (ICIs), and the positive rates of PD-L1 expression. The HR of PFS and OS and the merged ratios were calculated using a meta-analysis. The correlation between STK11mut and clinical characteristics was further analyzed in NSCLC datasets from public databases.ResultsFourteen retrospective studies including 4317 patients with NSCLC of whom 605 had STK11mut were included. The meta-analysis revealed that the ORR of ICIs in patients with STK11mut was 10.1% (95%CI 0.9–25.2), and the positive rate of PD-L1 expression was 41.1% (95%CI 25.3–57.0). STK11mut was associated with poor PFS (HR = 1.49, 95%CI 1.28–1.74) and poor OS (HR = 1.44, 95%CI 1.24–1.67). In the bioinformatics analysis, PFS and OS in patients with STK11 alterations were worse than those in patients without alterations (p < 0.001, p = 0.002). Nutlin-3a, 5-fluorouracil, and vinorelbine may have better sensitivity in patients with STK11mut than in those with STK11wt.ConclusionsPatients with STK11-mutant NSCLC had low PD-L1 expression and ORR to ICIs, and their PFS and OS were worse than patients with STK11wt after comprehensive treatment. In the future, more reasonable systematic treatments should be explored for this subgroup of patients with STK11-mutant NSCLC.

Abstract 887: Fast, interactive, AI-assisted 3D lung tumour segmentation

Abstract Segmentation of lung tumors is an important step in the planning of clinical treatment and operative procedures. Automatic segmentation of advanced lung tumors is challenging due to local invasion and heterogeneous metastasis. Fully automatic deep learning models often fail to accurately delineate advanced disease boundaries and do not allow the user to provide hints to improve segmentation performance. In contrast, interactive segmentation accepts user inputs and is an efficient way to segment lesions while reducing radiologist effort. This paper describes an interactive segmentation approach for lung lesion segmentation in CT scans. Our pipeline is built atop the SwinUNETR neural network architecture and can be expanded to support alternate models. The backbone model accepts radiologists' hints in the form of mouse clicks that indicate whether a location on the scan corresponds to a lesion. Our approach has the following novel contributions: (1) Context-aware encoding of guidance clicks by combining geodesic and Gaussian smoothing, resulting in improved segmentation of lesion boundaries. (2) An iterative prompting strategy to achieve higher accuracy with fewer clicks at inference time. (3) Guidance-aware Conditional Random Fields to refine the produced segmentation masks. (4) Volume cropping around guidance clicks at inference to improve segmentation precision and inference speed. (5) Linearly scaling our approach on multiple GPUs, to improve inference speed. On a public non-small cell lung cancer dataset predominantly containing advanced stage cancers (68\% stage III, and 27\% stage IV), our interactive approach delivers a Dice score of 0.74 with a single click, a 21\% improvement from the fully automatic approach (Dice Score = 0.61), and a Dice score of 0.81 with 5 clicks, which is a 33\% improvement. Inference speed is an important metric for an interactive pipeline. On a typical 350 $\times$ 350 $\times$ 150 voxel chest CT scan volume, our interactive approach infers 5 times faster than the original fully automated approach on a single GPU. After scaling to 4 GPUs, we infer in only 0.5 seconds; 20 times faster than the original approach. Qualitative evaluation from three independent radiologists indicates that our interactive pipeline significantly improved their standard clinical segmentation experience. Citation Format: Mayank Patwari, Yi Wei, Meng Xu, Zhenning Zhang, Konstantinos Sidiropoulos, Balaji Selvaraj, Georgia Hughes, Nadine Garibli, Kamsiriochukwu Ojiako, Mohan Lella, Leon Fedden, James Parkin, Michael Parker, Shaneil Patel, Qin Li, Kedar Patwardhan. Fast, interactive, AI-assisted 3D lung tumour segmentation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 887.

Regulatory role of the miR-142-3p/CDC25C axis in modulating autophagy in non-small cell lung cancer.

With its diverse genetic foundation and heterogeneous nature, non-small cell lung cancer (NSCLC) needs a better comprehension of prognostic evaluation and efficient treatment targeting. Bioinformatics analysis was performed of The Cancer Genome Atlas (TCGA)-NSCLC and GSE68571 dataset. Overlapping differentially expressed genes (DEGs) were used for functional enrichment analysis and constructing the protein-protein interaction (PPI) network. In addition, key prognostic genes were identified through prognostic risk models, and their expression levels were verified. The phenotypic effects of cell division cycle 25C (CDC25C) regulation on NSCLC cell lines were assessed by in vitro experiments using various techniques such as flow cytometry, Transwell, and colony formation. Protein levels related to autophagy and apoptosis were assessed, specifically examining the impact of autophagy inhibition [3-methyladenine (3-MA)] and the miR-142-3p/CDC25C axis on this regulatory system. CDC25C was identified as a key prognostic marker in NSCLC, showing high expression in tumor samples. In vitro experiments showed that CDC25C knockdown markedly reduced the capacity of cells to proliferate, migrate, invade, trigger apoptosis, and initiate cell cycle arrest. CDC25C and miR-142-3p displayed a reciprocal regulatory relationship. CDC25C reversed the inhibitory impacts of miR-142-3p on NSCLC cell cycle proliferation and progression. The synergy of miR-142-3p inhibition, CDC25C silencing, and 3-MA treatment was shown to regulate NSCLC cell processes including proliferation, apoptosis, and autophagy. MiR-142-3p emerged as a key player in governing autophagy and apoptosis by directly targeting CDC25C expression. This emphasizes the pivotal role of the miR-142-3p/CDC25C axis as a critical regulatory pathway in NSCLC.

A deep learning-based framework (Co-ReTr) for auto-segmentation of non-small cell-lung cancer in computed tomography images.

Deep learning-based auto-segmentation algorithms can improve clinical workflow by defining accurate regions of interest while reducing manual labor. Over the past decade, convolutional neural networks (CNNs) have become prominent in medical image segmentation applications. However, CNNs have limitations in learning long-range spatial dependencies due to the locality of the convolutional layers. Transformers were introduced to address this challenge. In transformers with self-attention mechanism, even the first layer of information processing makes connections between distant image locations. Our paper presents a novel framework that bridges these two unique techniques, CNNs and transformers, to segment the gross tumor volume (GTV) accurately and efficiently in computed tomography (CT) images of non-small cell-lung cancer (NSCLC) patients. Under this framework, input of multiple resolution images was used with multi-depth backbones to retain the benefits of high-resolution and low-resolution images in the deep learning architecture. Furthermore, a deformable transformer was utilized to learn the long-range dependency on the extracted features. To reduce computational complexity and to efficiently process multi-scale, multi-depth, high-resolution 3D images, this transformer pays attention to small key positions, which were identified by a self-attention mechanism. We evaluated the performance of the proposed framework on a NSCLC dataset which contains 563 training images and 113 test images. Our novel deep learning algorithm was benchmarked against five other similar deep learning models. The experimental results indicate that our proposed framework outperforms other CNN-based, transformer-based, and hybrid methods in terms of Dice score (0.92) and Hausdorff Distance (1.33). Therefore, our proposed model could potentially improve the efficiency of auto-segmentation of early-stage NSCLC during the clinical workflow. This type of framework may potentially facilitate online adaptive radiotherapy, where an efficient auto-segmentation workflow is required. Our deep learning framework, based on CNN and transformer, performs auto-segmentation efficiently and could potentially assist clinical radiotherapy workflow.

FunSpace: A functional and spatial analytic approach to cell imaging data using entropy measures.

Spatial heterogeneity in the tumor microenvironment (TME) plays a critical role in gaining insights into tumor development and progression. Conventional metrics typically capture the spatial differential between TME cellular patterns by either exploring the cell distributions in a pairwise fashion or aggregating the heterogeneity across multiple cell distributions without considering the spatial contribution. As such, none of the existing approaches has fully accounted for the simultaneous heterogeneity caused by both cellular diversity and spatial configurations of multiple cell categories. In this article, we propose an approach to leverage spatial entropy measures at multiple distance ranges to account for the spatial heterogeneity across different cellular organizations. Functional principal component analysis (FPCA) is applied to estimate FPC scores which are then served as predictors in a Cox regression model to investigate the impact of spatial heterogeneity in the TME on survival outcome, potentially adjusting for other confounders. Using a non-small cell lung cancer dataset (n = 153) as a case study, we found that the spatial heterogeneity in the TME cellular composition of CD14+ cells, CD19+ B cells, CD4+ and CD8+ T cells, and CK+ tumor cells, had a significant non-zero effect on the overall survival (p = 0.027). Furthermore, using a publicly available multiplexed ion beam imaging (MIBI) triple-negative breast cancer dataset (n = 33), our proposed method identified a significant impact of cellular interactions between tumor and immune cells on the overall survival (p = 0.046). In simulation studies under different spatial configurations, the proposed method demonstrated a high predictive power by accounting for both clinical effect and the impact of spatial heterogeneity.

Canagliflozin mediates tumor suppression alone and in combination with radiotherapy in non-small cell lung cancer (NSCLC) through inhibition of HIF-1α.

Non-small cell lung cancer (NSCLC) has a poor prognosis and effective therapeutic strategies are lacking. The diabetes drug canagliflozin inhibits NSCLC cell proliferation and the mammalian target of rapamycin (mTOR) pathway, which mediates cell growth and survival, but it is unclear whether this drug can enhance response rates when combined with cytotoxic therapy. Here, we evaluated the effects of canagliflozin on human NSCLC response to cytotoxic therapy in tissue cultures and xenografts. Ribonucleic acid sequencing (RNA-seq), real-time quantitative PCR (RT-qPCR), metabolic function, small interfering ribonucleic acid (siRNA) knockdown and protein expression assays were used in mechanistic analyses. We found that canagliflozin inhibited proliferation and clonogenic survival of NSCLC cells, and augmented the efficacy of radiotherapy to mediate these effects and inhibit NSCLC xenograft growth. Canagliflozin treatment alone moderately inhibited mitochondrial oxidative phosphorylation and exhibited greater anti-proliferative capacity than specific mitochondrial complex-I inhibitors. The treament downregulated genes mediating hypoxia-inducible factor (HIF)-1α stability, metabolism and survival, activated adenosine monophosphate-activated protein kinase (AMPK) and inhibited mTOR, a critical activator of HIF-1α signaling. HIF-1α knockdown and stabilization experiments suggested that canagliflozin mediates anti-proliferative effects, in part, through suppression of HIF-1α. Transcriptional regulatory network analysis pinpointed histone deacetylase 2 (HDAC2), a gene suppressed by canagliflozin, as a key mediator of canagliflozin's transcriptional reprogramming. HDAC2 knockdown eliminated HIF-1α levels and enhanced the anti-proliferative effects of canagliflozin. HDAC2-regulated genes suppressed by canagliflozin are associated with poor prognosis in several clinical NSCLC datasets. In addition, we include evidence that canagliflozin also improves NSCLC response to chemotherapy. In summary, canagliflozin may be a promising therapy to develop in combination with cytotoxic therapy in NSCLC.

A PI3K gene expression signature predicts for recurrence in early-stage non-small cell lung cancer treated with stereotactic body radiation therapy.

Increasingly, early-stage non-small cell lung cancer (NSCLC) is treated with stereotactic body radiation therapy (SBRT). Although treatment is generally effective, a small subset of tumors will recur because of radioresistance. Preclinical studies suggested PI3K-AKT-mTOR activation mediates radioresistance. This study sought to validate this finding in tumor samples from patients who underwent SBRT for NSCLC. Patients with T1-3N0 NSCLC treated with SBRT at our institution were included. Total RNA of formalin-fixed paraffin-embedded tumor biopsy specimens (pretherapy) was isolated and analyzed using the Clariom D assay. Risk scores from a PI3K activity signature and four published NSCLC signatures were generated and dichotomized by the median. Kaplan-Meier curves and Cox regressions were used to analyze their association with recurrence and overall survival (OS). The PI3K signature was also tested in a data set of resected NSCLC for additional validation. A total of 92 patients were included, with a median follow-up of 18.3months for living patients. There was no association of any of the four published gene expression signatures with recurrence or OS. However, high PI3K risk score was associated with higher local recurrence (hazard ratio [HR],11.72; 95% CI, 1.40-98.0; p=.023) and worse disease-free survival (DFS) (HR,3.98; 95% CI, 1.57-10.09; p=.0035), but not OS (p=.49), regional recurrence (p=.15), or distant recurrence (p=.85). In the resected NSCLC data set (n=361), high PI3K risk score was associated with decreased OS (log-rank p=.013) but not DFS (p=0.54). This study validates that higher PI3K activity, measured by gene expression, is associated with local recurrence and worse DFS in early-stage NSCLC patients treated with SBRT. This may be useful in prognostication and/or tailoring treatment, and merits further validation.

A risk score combining co-expression modules related to myeloid cells and alternative splicing associates with response to PD-1/PD-L1 blockade in non-small cell lung cancer.

Comprehensive analysis of transcriptomic profiles of non-small cell lung cancer (NSCLC) may provide novel evidence for biomarkers associated with response to PD-1/PD-L1 immune checkpoint blockade (ICB). We utilized weighted gene co-expression network analysis (WGCNA) to analyze transcriptomic data from two NSCLC datasets from Gene Expression Omnibus (GSE135222 and GSE126044) that involved patients received ICB treatment. We evaluated the correlation of co-expression modules with ICB responsiveness and functionally annotated ICB-related modules using pathway enrichment analysis, single-cell RNA sequencing, flow cytometry and alternative splicing analysis. We built a risk score using Lasso-COX regression based on hub genes from ICB-related modules. We investigated the alteration of tumor microenvironment between high- and low- risk groups and the association of the risk score with previously established predictive biomarkers. Our results identified a black with positive correlation and a blue module with negative correlation to ICB responsiveness. The black module was enriched in pathway of T cell activation and antigen processing and presentation, and the genes assigned to it were consistently expressed on myeloid cells. We observed decreased alternative splicing events in samples with high signature scores of the blue module. The Lasso-COX analysis screened out three genes (EVI2B, DHX9, HNRNPM) and constructed a risk score from the hub genes of the two modules. We validated the predictive value of the risk score for poor response to ICB therapy in an in-house NSCLC cohort and a pan-cancer cohort from the KM-plotter database. The low-risk group had more immune-infiltrated microenvironment, with higher frequencies of precursor exhausted CD8+ T cells, tissue-resident CD8+ T cells, plasmacytoid dendritic cells and type 1 conventional dendritic cells, and a lower frequency of terminal exhausted CD8+ T cells, which may explain its superior response to ICB therapy. The significant correlation of the risk score to gene signature of tertiary lymphoid structure also implicated the possible mechanism of this predictive biomarker. Our study identified two co-expression modules related to ICB responsiveness in NSCLC and developed a risk score accordingly, which could potentially serve as a predictive biomarker for ICB response.

Impact of CLEC3B mRNA expression on genetic landscape and immune tumor-microenvironment in NSCLC.

e20517 Background: Downregulation of C-type lectin domain family 3 member B (CLEC3B) is observed in non-small cell lung cancer (NSCLC), but its role remains largely unclear. Thus, we conducted this large scale genomic and transcriptomic analysis to gain novel insights into molecular and immunological functions of CLEC3B mRNA expression in NSCLC. Methods: Five publicly available WTS/WES NSCLC datasets (TCGA LUAD/LUSC, GSE41271, GSE66863, GSE81089, BATTLE) and the lung cancer single-cell atlas (Salcher, Cancer Cell 2022) were re-analyzed to decipher various molecular and biological aspects of CLEC3B in NSCLC. Findings were validated in a cohort of 19,982 NSCLC samples, which were centrally profiled (Caris Life Sciences, Phoenix, AZ) using WES/WTS. The cohort was stratified in quartiles according to CLEC3B mRNA expression status, with CLEC3Bhigh ( CLEC3BH) and CLEC3Blow ( CLEC3BL) expression defined as top and bottom quartile of transcripts per million (TPM) for further comparison. Immune cell fractions were calculated using QuantiSeq. Real-world overall survival (OS) was calculated using data from insurance claims. Results: CLEC3B mRNA expression was diminished in NSCLC compared to matched normal tissue (p &lt; 0.001). Adenomatous histology was associated with higher CLEC3B expression compared to squamous/large-cell carcinomas (p &lt; 0.005). In CLEC3B H tumors we observed lower rates of somatic mutations in TP53 (44.1 vs 81.7%), TTN (54.1 vs 79%), CDKN2A (5.2 vs 15%), but higher rates in KRAS (24 vs 7.8%) and STK11 (15.8 vs 2.69%) (all, q &lt; 0.01), which we corroborated in our validation cohort (TP53 (52.4 vs 74.7%), CDKN2A (8.3 vs 13%), RB1 (9.2 vs 13.5%), KMT2D (3.8 vs 7.3%), KRAS (31.8 vs 25.6%), EGFR (17.2 vs 8.1%) and STK11 (15.9 vs 11.3%, all q &lt; 0.05)). Further, CLEC3BH tumors were characterized by an immunosuppressive phenotype reflected by lower TMB and lower rates of MSI-H/dMMR, lower PD-L1 IHC expression and abundance of M2 macrophages and regulatory T-cells. Analysis of bulk and single cell transcriptomic datasets revealed that CLEC3BH was linked to endothelium-specific signaling and higher expression levels in endothelial cells. Finally, increased CLEC3B expression was associated with improved OS in the exploratory and validation cohorts. Conclusions: We herein described the molecular landscape of NSCLCs according to CLEC3B mRNA expression. We speculate that CLEC3B is mainly expressed by endothelial cells, and observed that CLEC3BH was associated with a distinct genetic profile and an immunosuppressive tumor-microenvironment. Thus, CLEC3B expression warrants further investigation as a stratification marker for NSCLC patients undergoing immune checkpoint therapy.

Artificial intelligence algorithm developed to predict immune checkpoint inhibitors efficacy in non–small-cell lung cancer.

9132 Background: Many non-small cell lung cancer (NSCLC) patients with high PD-L1 IHC expression and/or high TMB level do not respond to immune-check point inhibitors (ICIs). Discovery of novel predictive biomarkers for ICI response in NSCLC continues to be a critical unmet medical need, given the expense, potential toxicity, and therapeutic delay due to ICI treatment without benefit. Methods: We developed a supervised deep learning algorithm (Deep-IO) to predict therapeutic response to ICIs in NSCLC patients from standard histology whole slide hematoxylin and eosin (H&amp;E) images, which was trained based on ICI objective response rate. We utilized a convolutional neural network (CNN) implemented in Pytorch. The efficiency of the Deep-IO was evaluated through classification performance metrics and the area under the receiver operating characteristic curve (AUROC). The algorithm was trained and tested on 85218 tiles (size: 512x512 pixels) in 446 advanced stage NSCLC patients from the Dana Farber Cancer Institute (DFCI), all treated with ICI monotherapy. Results: The objective response rate, comprising complete and partial response, was 26% (n = 114) in the overall cohort. The DFCI patient cohort was randomly split into a training set (N=379, 85%) and a test set (N=67, 15%). The classifier’s predicted class (responder vs non-responder) probabilities at the tiles level were averaged for each patient. Quantitative evaluation of whole section test dataset shows that the developed model achieves the accuracy of 0.72, weighted average (WAVG) precision of 0.77, WAVG recall of 0.72 and WAVG F-score of 0.74 on classifying ICI non-responders from responders. Comparing with PD-L1 expression (TPS%) and TMB (mu/Mb), Deep-IO had superior predictive power for ICI response (AUROC=0.75) in the test set. The combined Deep-IO+PD-L1 scores resulted in a significantly greater AUROC value of 0.82 compared to single tests and a 32% improvement in specificity (0.88 vs. 0.56) compared to PD-L1 evaluation alone (Table). Conclusions: This proof-of-concept study shows that an artificial intelligence–powered H&amp;E image classifier can predict ICI effectiveness in NSCLC. Deep-IO outperformed both established predictive biomarkers. We will assess Deep-IO in additional external NSCLC data sets for validation and confirmation. [Table: see text]

PATZ1 in Non-Small Cell Lung Cancer: A New Biomarker That Negatively Correlates with PD-L1 Expression and Suppresses the Malignant Phenotype.

Non-small cell lung cancer (NSCLC), the leading cause of cancer death worldwide, is still an unmet medical problem due to the lack of both effective therapies against advanced stages and markers to allow a diagnosis of the disease at early stages before its progression. Immunotherapy targeting the PD-1/PD-L1 checkpoint is promising for many cancers, including NSCLC, but its success depends on the tumor expression of PD-L1. PATZ1 is an emerging cancer-related transcriptional regulator and diagnostic/prognostic biomarker in different malignant tumors, but its role in lung cancer is still obscure. Here we investigated expression and role of PATZ1 in NSCLC, in correlation with NSCLC subtypes and PD-L1 expression. A cohort of 104 NSCLCs, including lung squamous cell carcinomas (LUSCs) and adenocarcinomas (LUADs), was retrospectively analyzed by immunohistochemistry for the expression of PATZ1 and PD-L1. The results were correlated with each other and with the clinical characteristics, showing on the one hand a positive correlation between the high expression of PATZ1 and the LUSC subtype and, on the other hand, a negative correlation between PATZ1 and PD-L1, validated at the mRNA level in independent NSCLC datasets. Consistently, two NSCLC cell lines transfected with a PATZ1-overexpressing plasmid showed PD-L1 downregulation, suggesting a role for PATZ1 in the negative regulation of PD-L1. We also showed that PATZ1 overexpression inhibits NSCLC cell proliferation, migration, and invasion, and that Patz1-knockout mice develop LUAD. Overall, this suggests that PATZ1 may act as a tumor suppressor in NSCLC.

Abstract 4396: TROP2 expression in non-small cell lung cancer

Abstract Introduction: Non-small cell lung cancer (NSCLC) is a major cause of cancer-related deaths globally with high unmet need. A variety of novel therapeutic strategies are being explored to improve patient outcomes including the use of antibody drug conjugates (ADCs) to deliver highly potent, cytotoxic payloads to tumors. Trophoblast cell-surface antigen 2 (TROP2), also known as tumor-associated calcium signal transducer 2 (TACSTD2), has emerged as an attractive target for ADCs. Sacituzumab govitecan-hziy is a novel ADC composed of a TROP2 antibody coupled to SN-38 and is being evaluated in NSCLC. High TROP2 mRNA expression is observed in many tumor types, however, TROP2 protein expression in NSCLC is not well-established. Here we characterized TROP2 expression across three independent datasets to explore the TROP2 relationship to baseline characteristics, molecular features of interest, and its prognostic value in NSCLC. Methods: We analyzed three independent datasets: (1) The Cancer Genome Atlas (TCGA) NSCLC adenocarcinoma (LUAD, N=660) and squamous cell carcinoma (LUSC, N=484) mRNA expression data, (2) a NSCLC adenocarcinoma FFPE tumor sample set (“Translational data set”; N=107), (3) an independent clinical NSCLC adenocarcinoma (N=103) and squamous cell carcinoma (N=37) data set. TROP2 IHC was performed on tumor sample sets (2) and (3) using the SP295 antibody (Robust Prototype Assay, Ventana) with H-Score and assessment of the percentage of membrane-positive tumor cells as readouts. Wilcoxon rank-sum and Kruskal-Wallis tests (continuous variables), Kaplan-Meier method (survival probability curves) and log-rank test (time to event outcome) were used (GraphPad Prism 8.1.2 and R version 4.0.5). Results: TROP2 mRNA was highly expressed in NSCLC and expression was similar in adenocarcinoma and squamous cell carcinoma in the TCGA data set. TROP2 expression did not vary with patient’s age, gender, race or tumor stage. Furthermore, TROP2 expression was not associated with TP53, KRAS or driver mutation status (including EGFR, ALK, ROS, RET, MET). TROP2 expression in the Translational sample set was consistent with the findings from the TCGA data analysis: TROP2 mRNA and protein expression were high and independent of patient baseline demographics, tumor stage and TP53 and KRAS mutation status. Finally, in an independent clinical NSCLC dataset, TROP2 protein was highly expressed and there was no relationship between TROP2 and baseline demographics including gender and age. TROP2 membrane H-scores and percentage of membrane-positive tumor cells also highly correlated in the two independent datasets. Conclusions: Three large independent datasets confirmed high expression of TROP2 in NSCLC both, as mRNA and protein. TROP2 expression in tumors did not differ between histological subtypes, baseline characteristics or clinically relevant driver alterations. Based on the TCGA data set, TROP2 expression was not prognostic for survival. Citation Format: Peiwen Kuo, Emon Elboudwarej, Lauri Diehl, Jilpa Patel, Sabeen Mekan, Juliane M. Jürgensmeier. TROP2 expression in non-small cell lung cancer. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4396.

Synergies of Radiomics and Transcriptomics in Lung Cancer Diagnosis: A Pilot Study.

Radiotranscriptomics is an emerging field that aims to investigate the relationships between the radiomic features extracted from medical images and gene expression profiles that contribute in the diagnosis, treatment planning, and prognosis of cancer. This study proposes a methodological framework for the investigation of these associations with application on non-small-cell lung cancer (NSCLC). Six publicly available NSCLC datasets with transcriptomics data were used to derive and validate a transcriptomic signature for its ability to differentiate between cancer and non-malignant lung tissue. A publicly available dataset of 24 NSCLC-diagnosed patients, with both transcriptomic and imaging data, was used for the joint radiotranscriptomic analysis. For each patient, 749 Computed Tomography (CT) radiomic features were extracted and the corresponding transcriptomics data were provided through DNA microarrays. The radiomic features were clustered using the iterative K-means algorithm resulting in 77 homogeneous clusters, represented by meta-radiomic features. The most significant differentially expressed genes (DEGs) were selected by performing Significance Analysis of Microarrays (SAM) and 2-fold change. The interactions among the CT imaging features and the selected DEGs were investigated using SAM and a Spearman rank correlation test with a False Discovery Rate (FDR) of 5%, leading to the extraction of 73 DEGs significantly correlated with radiomic features. These genes were used to produce predictive models of the meta-radiomics features, defined as p-metaomics features, by performing Lasso regression. Of the 77 meta-radiomic features, 51 can be modeled in terms of the transcriptomic signature. These significant radiotranscriptomics relationships form a reliable basis to biologically justify the radiomics features extracted from anatomic imaging modalities. Thus, the biological value of these radiomic features was justified via enrichment analysis on their transcriptomics-based regression models, revealing closely associated biological processes and pathways. Overall, the proposed methodological framework provides joint radiotranscriptomics markers and models to support the connection and complementarities between the transcriptome and the phenotype in cancer, as demonstrated in the case of NSCLC.

Five-hub genes identify potential mechanisms for the progression of asthma to lung cancer.

Previous studies have shown that asthma is a risk factor for lung cancer, while the mechanisms involved remain unclear. We attempted to further explore the association between asthma and non-small cell lung cancer (NSCLC) via bioinformatics analysis. We obtained GSE143303 and GSE18842 from the GEO database. Lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) groups were downloaded from the TCGA database. Based on the results of differentially expressed genes (DEGs) between asthma and NSCLC, we determined common DEGs by constructing a Venn diagram. Enrichment analysis was used to explore the common pathways of asthma and NSCLC. A protein-protein interaction (PPI) network was constructed to screen hub genes. KM survival analysis was performed to screen prognostic genes in the LUAD and LUSC groups. A Cox model was constructed based on hub genes and validated internally and externally. Tumor Immune Estimation Resource (TIMER) was used to evaluate the association of prognostic gene models with the tumor microenvironment (TME) and immune cell infiltration. Nomogram model was constructed by combining prognostic genes and clinical features. 114 common DEGs were obtained based on asthma and NSCLC data, and enrichment analysis showed that significant enrichment pathways mainly focused on inflammatory pathways. Screening of 5 hub genes as a key prognostic gene model for asthma progression to LUAD, and internal and external validation led to consistent conclusions. In addition, the risk score of the 5 hub genes could be used as a tool to assess the TME and immune cell infiltration. The nomogram model constructed by combining the 5 hub genes with clinical features was accurate for LUAD. Five-hub genes enrich our understanding of the potential mechanisms by which asthma contributes to the increased risk of lung cancer.

Immune infiltration patterns and identification of new diagnostic biomarkers GDF10, NCKAP5, and RTKN2 in non-small cell lung cancer

This study aimed to identify potential biomarkers for non-small cell lung cancer (NSCLC) and analyze the role of immune cell infiltration in NSCLC. R software was used to screen differentially expressed genes (DEGs) from NSCLC datasets obtained from the Gene Expression Omnibus (GEO) database, and functional correlation analysis was performed. The machine learning algorithms were used to screen the potential biomarkers of NSCLC. The diagnostic values were assessed through receiver operating characteristic (ROC) curves. The protein and mRNA expression levels of potential biomarkers were verified based on the Human Protein Atlas (HPA) database and qRT-PCR. CIBERSORT was used to evaluate the infiltration of immune cells in NSCLC tissues, and the correlation between potential biomarkers and infiltrated immune cell was analyzed. Finally, specific siRNAs were utilized to reduce the GDF10, NCKAP5, and RTKN2 expression in A549 and H1975 cells. The proliferation ability of A549 and H1975 cells was detected by MTT assay. A total of 848 upregulated DEGs and 1308 downregulated DEGs were identified. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses showed that the DEGs were mainly related to cell division. Disease ontology (DO) enrichment analysis showed that the diseases with these DEGs were mainly lung diseases, including NSCLC. In addition,three potential biomarkers were identified: GDF10, NCKAP5, and RTKN2. Immune cell infiltration analysis showed that resting NK cells, activated dendritic cells, and Tregs may be involved in the pathogenesis of NSCLC. Meanwhile, GDF10, NCKAP5, and RTKN2 were negatively correlated with Tregs and naïve B cells but were positively correlated with activated dendritic cells and resting NK cells. Immunohistochemical staining showed that the expression of GDF10, NCKAP5, and RTKN2 in the lung tissue of patients with NSCLC was lower than that of normal lung tissue. qRT-PCR also confirmed that the mRNA expression of three biomarkers in NSCLC cell lines A549 and H1975 were significantly lower than those in human normal lung epithelial cells BEAS-2B. An MTT assay showed that GDF10, NCKAP5, and RTKN2 knockdown significantly promoted the proliferation of A549 and H1975 cells. The in vitro experiments showed that GDF10, NCKAP5, and RTKN2 played the inhibitory effects on NSCLC cell lines proliferation. Hence, GDF10, NCKAP5, and RTKN2 can be used as diagnostic biomarkers for NSCLC.

Augmentation of predictive competence of non-small cell lung cancer datasets through feature pre-processing techniques

Augmentation of predictive competence of non-small cell lung cancer datasets through feature pre-processing techniques

Augmentation of Predictive Competence of Non-Small Cell Lung Cancer Datasets through Feature Pre-Processing Techniques

The major Objective of the Study is to augment the predictive analytics of Non-Small Cell Lung Cancer (NSCLC) datasets with Feature Pre-Processing (FPP) technique in three stages viz. Remove base errors with common analytics on emptiness or non-numerical or missing values in the dataset, remove repeated features through regression analysis and eliminate irrelevant features through clustering methods. The FPP Model is validated using classifiers like simple and complex Tree, Linear and Gaussian SVM, Weighted KNN and Boosted Trees in terms of accuracy, sensitivity, specificity, kappa, positive and negative likelihood. The result showed that the NSCLC dataset formed after FPP outperformed the raw NSCLC dataset in all performance levels and showed good augmentation in predictive analytics of NSCLC datasets. The research proved that preprocessing is essential for better prediction of complex medical datasets.

Towards survival prediction of cancer patients using medical images.

Survival prediction of a patient is a critical task in clinical medicine for physicians and patients to make an informed decision. Several survival and risk scoring methods have been developed to estimate the survival score of patients using clinical information. For instance, the Global Registry of Acute Coronary Events (GRACE) and Thrombolysis in Myocardial Infarction (TIMI) risk scores are developed for the survival prediction of heart patients. Recently, state-of-the-art medical imaging and analysis techniques have paved the way for survival prediction of cancer patients by understanding key features extracted from Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scanned images with the help of image processing and machine learning techniques. However, survival prediction is a challenging task due to the complexity in benchmarking of image features, feature selection methods, and machine learning models. In this article, we evaluate the performance of 156 visual features from radiomic and hand-crafted feature classes, six feature selection methods, and 10 machine learning models to benchmark their performance. In addition, MRI scanned Brain Tumor Segmentation (BraTS) and CT scanned non-small cell lung cancer (NSCLC) datasets are used to train classification and regression models. Our results highlight that logistic regression outperforms for the classification with 66 and 54% accuracy for BraTS and NSCLC datasets, respectively. Moreover, our analysis of best-performing features shows that age is a common and significant feature for survival prediction. Also, gray level and shape-based features play a vital role in regression. We believe that the study can be helpful for oncologists, radiologists, and medical imaging researchers to understand and automate the procedure of decision-making and prognosis of cancer patients.

De novo serine biosynthesis from glucose predicts sex-specific response to antifolates in non-small cell lung cancer cell lines

Lung cancer is the leading cause of cancer-related death. Intriguingly, males with non-small cell lung cancer (NSCLC) have a higher mortality rate than females. Here, we investigated the role of serine metabolism as a predictive marker for sensitivity to the antifolate pemetrexed in male and female NSCLC cell lines. Using [13C6] glucose tracing in NSCLC cell lines, we found that a subset of male cells generated significantly more serine from glucose than female cells. Higher serine biosynthesis was further correlated with increased sensitivity to pemetrexed in male cells only. Concordant sex differences in metabolic gene expression were evident in NSCLC and pan-cancer transcriptome datasets, suggesting a potential mechanism with wide-reaching applicability. These data were further validated by integrating antifolate drug cytotoxicity and metabolic pathway transcriptome data from pan-cancer cell lines. Together, these findings highlight the importance of considering sex differences in cancer metabolism to improve treatment for all patients.

Reconstruction-Assisted Feature Encoding Network for Histologic Subtype Classification of Non-Small Cell Lung Cancer.

Accurate histological subtype classification between adenocarcinoma (ADC) and squamous cell carcinoma (SCC) using computed tomography (CT) images is of great importance to assist clinicians in determining treatment and therapy plans for non-small cell lung cancer (NSCLC) patients. Although current deep learning approaches have achieved promising progress in this field, they are often difficult to capture efficient tumor representations due to inadequate training data, and in consequence show limited performance. In this study, we propose a novel and effective reconstruction-assisted feature encoding network (RAFENet) for histological subtype classification by leveraging an auxiliary image reconstruction task to enable extra guidance and regularization for enhanced tumor feature representations. Different from existing reconstruction-assisted methods that directly use generalizable features obtained from shared encoder for primary task, a dedicated task-aware encoding module is utilized in RAFENet to perform refinement of generalizable features. Specifically, a cascade of cross-level non-local blocks are introduced to progressively refine generalizable features at different levels with the aid of lower-level task-specific information, which can successfully learn multi-level task-specific features tailored to histological subtype classification. Moreover, in addition to widely adopted pixel-wise reconstruction loss, we introduce a powerful semantic consistency loss function to explicitly supervise the training of RAFENet, which combines both feature consistency loss and prediction consistency loss to ensure semantic invariance during image reconstruction. Extensive experimental results show that RAFENet effectively addresses the difficult issues that cannot be resolved by existing reconstruction-based methods and consistently outperforms other state-of-the-art methods on both public and in-house NSCLC datasets. Supplementary material is available at https://github.com/lhch1994/Rafenet_sup_material.