Gene Expression Classification of Lung Adenocarcinoma into Molecular Subtypes

Role of AI in empowering and redefining the oncology care landscape: perspective from a developing nation

Early diagnosis and accurate prognosis play a pivotal role in the clinical management of cancer and in preventing cancer-related mortalities. The burgeoning population of Asia in general and South Asian countries like India in particular pose significant challenges to the healthcare system. Regrettably, the demand for healthcare services in India far exceeds the available resources, resulting in overcrowded hospitals, prolonged wait times, and inadequate facilities. The scarcity of trained manpower in rural settings, lack of awareness and low penetrance of screening programs further compounded the problem. Artificial Intelligence (AI), driven by advancements in machine learning, deep learning, and natural language processing, can profoundly transform the underlying shortcomings in the healthcare industry, more for populous nations like India. With about 1.4 million cancer cases reported annually and 0.9 million deaths, India has a significant cancer burden that surpassed several nations. Further, India's diverse and large ethnic population is a data goldmine for healthcare research. Under these circumstances, AI-assisted technology, coupled with digital health solutions, could support effective oncology care and reduce the economic burden of GDP loss in terms of years of potential productive life lost (YPPLL) due to India's stupendous cancer burden. This review explores different aspects of cancer management, such as prevention, diagnosis, precision treatment, prognosis, and drug discovery, where AI has demonstrated promising clinical results. By harnessing the capabilities of AI in oncology research, healthcare professionals can enhance their ability to diagnose cancers at earlier stages, leading to more effective treatments and improved patient outcomes. With continued research and development, AI and digital health can play a transformative role in mitigating the challenges posed by the growing population and advancing the fight against cancer in India. Moreover, AI-driven technologies can assist in tailoring personalized treatment plans, optimizing therapeutic strategies, and supporting oncologists in making well-informed decisions. However, it is essential to ensure responsible implementation and address potential ethical and privacy concerns associated with using AI in healthcare.

Unveiling the impact of cell death-related genes and immune dynamics on drug resistance in lung adenocarcinoma: a risk score model and functional insights

Lung adenocarcinoma (LUAD), characterized by its heterogeneity and complex pathogenesis, is the focus of this study which investigates the association between cell death-related genes and LUAD. Through machine learning, a risk score model was developed using the Coxboost rsf algorithm, demonstrating strong prognostic accuracy in both validation (GSE30219, GSE31210, GSE72094) and training (TCGA-LUAD) datasets with C-indices of 0.93, 0.67, 0.68, and 0.64, respectively. The study reveals that the expression of Keratin 18 (KRT18), a key cytoskeletal protein, varies across LUAD cell lines (DV-90, PC-9, A549) compared to normal bronchial epithelial cells (BEAS-2B), suggesting its potential role in LUAD's pathogenesis. Kaplan-Meier survival curves further validate the model, indicating longer survival in the low-risk group. A comprehensive analysis of gene expression, functional differences, immune infiltration, and mutations underscores significant variations between risk groups, highlighting the high-risk group's immunological dysfunction. This points to a more intricate tumor immune environment and the possibility of alternative therapeutic strategies. The study also delves into drug sensitivity, showing distinct responses between risk groups, underscoring the importance of risk stratification in treatment decisions for LUAD patients. Additionally, it explores KRT18's epigenetic regulation and its correlation with immune cell infiltration and immune regulatory molecules, suggesting KRT18's significant role in the tumor immune landscape. This research not only offers a valuable prognostic tool for LUAD but also illuminates the complex interplay between cell death-related genes, drug sensitivity, and immune infiltration, positioning KRT18 as a potential therapeutic or prognostic target to improve patient outcomes by personalizing LUAD treatment strategies.

Machine learning enabled classification of lung cancer cell lines co-cultured with fibroblasts with lightweight convolutional neural network for initial diagnosis

BACKGROUND

Identification of lung cancer subtypes is critical for successful treatment in patients, especially those in advanced stages. Many advanced and personal treatments require knowledge of specific mutations, as well as up- and down-regulations of genes, for effective targeting of the cancer cells. While many studies focus on individual cell structures and delve deeper into gene sequencing, the present study proposes a machine learning method for lung cancer classification based on low-magnification cancer outgrowth patterns in a 2D co-culture environment.

METHODS

Using a magnetic well plate holder, circular pattern lung cancer cell clusters were generated among fibroblasts, and daily images were captured to monitor cancer outgrowth over a 9-day period. These outgrowth images were then augmented and used to train a convolutional neural network (CNN) model based on the lightweight TinyVGG architecture. The model was trained with pairs of classes representing three subtypes of NSCLC: A549 (adenocarcinoma), H520 (squamous cell carcinoma), and H460 (large cell carcinoma). The objective was to assess whether this lightweight machine learning model could accurately classify the three lung cancer cell lines at different stages of cancer outgrowth. Additionally, cancer outgrowth images of two patient-derived lung cancer cells, one with the KRAS oncogene and the other with the EGFR oncogene, were captured and classified using the CNN model. This demonstration aimed to investigate the translational potential of machine learning-enabled lung cancer classification.

RESULTS

The lightweight CNN model achieved over 93% classification accuracy at 1 day of outgrowth among A549, H460, and H520, and reached 100% classification accuracy at 7 days of outgrowth. Additionally, the model achieved 100% classification accuracy at 4 days for patient-derived lung cancer cells. Although these cells are classified as Adenocarcinoma, their outgrowth patterns vary depending on their oncogene expressions (KRAS or EGFR).

CONCLUSIONS

These results demonstrate that the lightweight CNN architecture, operating locally on a laptop without network or cloud connectivity, can effectively create a machine learning-enabled model capable of accurately classifying lung cancer cell subtypes, including those derived from patients, based upon their outgrowth patterns in the presence of surrounding fibroblasts. This advancement underscores the potential of machine learning to enhance early lung cancer subtyping, offering promising avenues for improving treatment outcomes in advanced stage-patients.

Establishing a metastasis-related diagnosis and prognosis model for lung adenocarcinoma through CRISPR library and TCGA database

PURPOSE

Existing biomarkers for diagnosing and predicting metastasis of lung adenocarcinoma (LUAD) may not meet the demands of clinical practice. Risk prediction models with multiple markers may provide better prognostic factors for accurate diagnosis and prediction of metastatic LUAD.

METHODS

An animal model of LUAD metastasis was constructed using CRISPR technology, and genes related to LUAD metastasis were screened by mRNA sequencing of normal and metastatic tissues. The immune characteristics of different subtypes were analyzed, and differentially expressed genes were subjected to survival and Cox regression analyses to identify the specific genes involved in metastasis for constructing a prediction model. The biological function of RFLNA was verified by analyzing CCK-8, migration, invasion, and apoptosis in LUAD cell lines.

RESULTS

We identified 108 differentially expressed genes related to metastasis and classified LUAD samples into two subtypes according to gene expression. Subsequently, a prediction model composed of eight metastasis-related genes (RHOBTB2, KIAA1524, CENPW, DEPDC1, RFLNA, COL7A1, MMP12, and HOXB9) was constructed. The areas under the curves of the logistic regression and neural network were 0.946 and 0.856, respectively. The model effectively classified patients into low- and high-risk groups. The low-risk group had a better prognosis in both the training and test cohorts, indicating that the prediction model had good diagnostic and predictive power. Upregulation of RFLNA successfully promoted cell proliferation, migration, invasion, and attenuated apoptosis, suggesting that RFLNA plays a role in promoting LUAD development and metastasis.

CONCLUSION

The model has important diagnostic and prognostic value for metastatic LUAD and may be useful in clinical applications.

Dual Polarization Modality Fusion Network for Assisting Pathological Diagnosis

Polarization imaging is sensitive to sub-wavelength microstructures of various cancer tissues, providing abundant optical characteristics and microstructure information of complex pathological specimens. However, how to reasonably utilize polarization information to strengthen pathological diagnosis ability remains a challenging issue. In order to take full advantage of pathological image information and polarization features of samples, we propose a dual polarization modality fusion network (DPMFNet), which consists of a multi-stream CNN structure and a switched attention fusion module for complementarily aggregating the features from different modality images. Our proposed switched attention mechanism could obtain the joint feature embeddings by switching the attention map of different modality images to improve their semantic relatedness. By including a dual-polarization contrastive training scheme, our method can synthesize and align the interaction and representation of two polarization features. Experimental evaluations on three cancer datasets show the superiority of our method in assisting pathological diagnosis, especially in small datasets and low imaging resolution cases. Grad-CAM visualizes the important regions of the pathological images and the polarization images, indicating that the two modalities play different roles and allow us to give insightful corresponding explanations and analysis on cancer diagnosis conducted by the DPMFNet. This technique has potential to facilitate the performance of pathological aided diagnosis and broaden the current digital pathology boundary based on pathological image features.

Deep learning-based prediction of molecular cancer biomarkers from tissue slides: A new tool for precision oncology

Molecular tests are necessary to stratify cancer patients for targeted therapy. However, high cost and technical barriers limit the application of these tests, hindering optimal treatment. Recently, deep learning (DL) has been applied to predict molecular test results from digitized images of tissue slides. Furthermore, treatment response and prognosis can be predicted from tissue slides using DL. In this review, we summarized DL-based studies regarding the prediction of genetic mutation, microsatellite instability, tumor mutational burden, molecular subtypes, gene expression, treatment response, and prognosis directly from hematoxylin- and eosin-stained tissue slides. Although performance needs to be improved, these studies clearly demonstrated the feasibility of DL-based prediction of key molecular features in cancer tissues. With the accumulation of data and technical advances, the performance of the DL system could be improved in the near future. Therefore, we expect that DL could provide cost- and time-effective alternative tools for patient stratification in the era of precision oncology.

Lightweight Deep Learning Classification Model for Identifying Low-Resolution CT Images of Lung Cancer

With an astounding five million fatal cases every year, lung cancer is among the leading causes of mortality worldwide for both men and women. The diagnosis of lung illnesses can benefit from the information a computed tomography (CT) scan can offer. The major goals of this study are to diagnose lung cancer and its seriousness and to identify malignant lung nodules from the provided input lung picture. This paper applies unique deep learning techniques to identify the exact location of the malignant lung nodules. Using a DenseNet model, mixed ground glass is analyzed in low-dose, low-resolution CT scan images of nodules (mGGNs) with a slice thickness of 5 mm in this study. This was done to categorize and identify many histological subtypes of lung cancer. Low-resolution CT scans are used to pathologically classify invasive adenocarcinoma (IAC) and minimally invasive adenocarcinoma (MIA). 105 low-resolution CT images with 5 mm thick slices from 105 patients at Lishui Central Hospital were selected. To detect and distinguish, IAC and MIA, extend and enhance deep learning two- and three-dimensional DenseNet models are used. The two-dimensional DenseNet model was shown to perform much better than the three-dimensional DenseNet model in terms of classification accuracy (76.67%), sensitivity (63.3%), specificity (100%), and area under the receiver operating characteristic curve (0.88). Finding the histological subtypes of persons with lung cancer should aid doctors in making a more precise diagnosis, even if the image quality is not outstanding.

Multi-Omics Integrative Analysis of Lung Adenocarcinoma: An in silico Profiling for Precise Medicine

Lung adenocarcinoma (LUAD) is one of the most common histological subtypes of lung cancer. The aim of this study was to construct consensus clusters based on multi-omics data and multiple algorithms. In order to identify specific molecular characteristics and facilitate the use of precision medicine on patients we used gene expression, DNA methylation, gene mutations, copy number variation data, and clinical data of LUAD patients for clustering. Consensus clusters were obtained using a consensus ensemble of five multi-omics integrative algorithms. Four molecular subtypes were identified. The CS1 and CS2 subtypes had better prognosis. Based on the immune and drug sensitivity predictions, we inferred that CS1 may be less responsive to immunotherapy and less sensitive to chemotherapeutic drugs. The high immune infiltration of CS2 cells may respond well to immunotherapy. Additionally, the CS2 subtype may also respond to EGFR molecular targeted therapy. The CS3 and CS4 subtypes were associated with poor prognosis. These two subtypes had more mutations, especially TP53 ones, as well as higher sensitivity to chemotherapeutics for lung cancer. However, CS3 was enriched in immune-related pathways and may respond to anti-PD1 immunotherapy. In addition, CS1 and CS4 were less sensitive to ferroptosis inhibitors. We performed a comprehensive analysis of the five types of omics data using five clustering algorithms to reveal the molecular characteristics of LUAD patients. These findings provide new insights into LUAD subtypes and potential clinical treatment strategies to guide personalized management and treatment.

Deep learning in cancer diagnosis, prognosis and treatment selection

Deep learning is a subdiscipline of artificial intelligence that uses a machine learning technique called artificial neural networks to extract patterns and make predictions from large data sets. The increasing adoption of deep learning across healthcare domains together with the availability of highly characterised cancer datasets has accelerated research into the utility of deep learning in the analysis of the complex biology of cancer. While early results are promising, this is a rapidly evolving field with new knowledge emerging in both cancer biology and deep learning. In this review, we provide an overview of emerging deep learning techniques and how they are being applied to oncology. We focus on the deep learning applications for omics data types, including genomic, methylation and transcriptomic data, as well as histopathology-based genomic inference, and provide perspectives on how the different data types can be integrated to develop decision support tools. We provide specific examples of how deep learning may be applied in cancer diagnosis, prognosis and treatment management. We also assess the current limitations and challenges for the application of deep learning in precision oncology, including the lack of phenotypically rich data and the need for more explainable deep learning models. Finally, we conclude with a discussion of how current obstacles can be overcome to enable future clinical utilisation of deep learning.

Gene Expression Profiling as a Potential Tool for Precision Oncology in Non-Small Cell Lung Cancer

Recent technological advances and the application of high-throughput mutation and transcriptome analyses have improved our understanding of cancer diseases, including non-small cell lung cancer. For instance, genomic profiling has allowed the identification of mutational events which can be treated with specific agents. However, detection of DNA alterations does not fully recapitulate the complexity of the disease and it does not allow selection of patients that benefit from chemo- or immunotherapy. In this context, transcriptional profiling has emerged as a promising tool for patient stratification and treatment guidance. For instance, transcriptional profiling has proven to be especially useful in the context of acquired resistance to targeted therapies and patients lacking targetable genomic alterations. Moreover, the comprehensive characterization of the expression level of the different pathways and genes involved in tumor progression is likely to better predict clinical benefit from different treatments than single biomarkers such as PD-L1 or tumor mutational burden in the case of immunotherapy. However, intrinsic technical and analytical limitations have hindered the use of these expression signatures in the clinical setting. In this review, we will focus on the data reported on molecular classification of non-small cell lung cancer and discuss the potential of transcriptional profiling as a predictor of survival and as a patient stratification tool to further personalize treatments.

Functional Roles of SPINK1 in Cancers

Serine Peptidase Inhibitor Kazal Type 1 (SPINK1) is a secreted protein known as a protease inhibitor of trypsin in the pancreas. However, emerging evidence shows its function in promoting cancer progression in various types of cancer. SPINK1 modulated tumor malignancies and induced the activation of the downstream signaling of epidermal growth factor receptor (EGFR) in cancer cells, due to the structural similarity with epidermal growth factor (EGF). The discoverable SPINK1 somatic mutations, expressional signatures, and prognostic significances in various types of cancer have attracted attention as a cancer biomarker in clinical applications. Emerging findings further clarify the direct and indirect biological effects of SPINK1 in regulating cancer proliferation, metastasis, drug resistance, transdifferentiation, and cancer stemness, warranting the exploration of the SPINK1-mediated molecular mechanism to identify a therapeutic strategy. In this review article, we first integrate the transcriptomic data of different types of cancer with clinical information and recent findings of SPINK1-mediated malignant phenotypes. In addition, a comprehensive summary of SPINK1 expression in a pan-cancer panel and individual cell types of specific organs at the single-cell level is presented to indicate the potential sites of tumorigenesis, which has not yet been reported. This review aims to shed light on the roles of SPINK1 in cancer and provide guidance and potential directions for scientists in this field.

Identification of 5-Gene Signature Improves Lung Adenocarcinoma Prognostic Stratification Based on Differential Expression Invasion Genes of Molecular Subtypes

Background

The acquisition of invasive tumor cell behavior is considered to be the cornerstone of the metastasis cascade. Thus, genetic markers associated with invasiveness can be stratified according to patient prognosis. In this study, we aimed to identify an invasive genetic trait and study its biological relevance in lung adenocarcinoma.

Methods

250 TCGA patients with lung adenocarcinoma were used as the training set, and the remaining 250 TCGA patients, 500 ALL TCGA patients, 226 patients with GSE31210, 83 patients with GSE30219, and 127 patients with GSE50081 were used as the verification data sets. Subtype classification of all TCGA lung adenocarcinoma samples was based on invasion-associated genes using the R package ConsensusClusterPlus. Kaplan-Meier curves, LASSO (least absolute contraction and selection operator) method, and univariate and multivariate Cox analysis were used to develop a molecular model for predicting survival.

Results

As a consequence, two molecular subtypes for LUAD were first identified from all TCGA all data sets which were significant on survival time. C1 subtype with poor prognosis has higher clinical characteristics of malignancy, higher mutation frequency of KRAS and TP53, and a lower expression of immune regulatory molecules. 2463 differentially expressed invasion genes between C1 and C2 subtypes were obtained, including 580 upregulation genes and 1883 downregulation genes. Functional enrichment analysis found that upregulated genes were associated with the development of tumor pathways, while downregulated genes were more associated with immunity. Furthermore, 5-invasion gene signature was constructed based on 2463 genes, which was validated in four data sets. This signature divided patients into high-risk and low-risk groups, and the LUDA survival rate of the high-risk group is significantly lower than that of the low-risk group. Multivariate Cox analysis revealed that this gene signature was an independent prognostic factor for LUDA. Compared with other existing models, our model has a higher AUC.

Conclusion

In this study, two subtypes were identified. In addition, we developed a 5-gene signature prognostic risk model, which has a good AUC in the training set and independent validation set and is a model with independent clinical characteristics. Therefore, we recommend using this classifier as a molecular diagnostic test to assess the prognostic risk of patients with LUDA.

Involvement of CHP2 in the Development of Non-Small Cell Lung Cancer and Patients' Poor Prognosis

The present study aimed to investigate the expression levels and clinical significance of the calcineurin B homologous protein 2 (CHP2) in non-small cell lung cancer (NSCLC), and to study its effects on biological characteristics of NSCLC cells. Tumor and adjacent samples were collected from 196 NSCLC patients. Western blot analysis was used to detect the expression levels of the CHP2 in 8 pairs of NSCLC fresh tissues and 4 NSCLC cell lines. Immunohistochemical analysis was used to detect the expression of the CHP2 in 188 additional pairs of NSCLC wax block tissues. The data indicated that the expression levels of the CHP2 in the paraffin and fresh tissues of NSCLC were significantly higher than those of the adjacent tissues. According to the histo-score, univariate and multivariate analysis indicated that a high expression level of CHP2 was an important factor affecting the 5-year survival rate of NSCLC patients. After knocking down the expression of CHP2 in NSCLC cell lines, the proliferative, migratory, and invasive activities of NSCLC-CHP2 cells were decreased which were assessed by Western blotting, Cell Counting Kit-8, and transwell and wound-healing assays. In conclusion, the data demonstrated that CHP2 was highly expressed in NSCLC and that it could promote the development of NSCLC, suggesting its potential application for the therapy of NSCLC.

Genomic profiling reveals the pivotal role of hrHPV driving copy number and gene expression alterations, including mRNA downregulation of TP53 and RB1 in penile cancer

The incidence of penile cancer (PeCa) is increasing worldwide, however, the highest rates are reported in underdeveloped countries. The molecular mechanisms that underly the onset and progression of these tumors are still unclear. Therefore, our goal was to determine the genome-wide copy number alterations and the involvement of human papiloma virus (HPV) (TP53 and RB1), inflammatory (COX2 and EGFR), and PI3K/AKT pathway (AKT1, AKT2, EGFR, ERBB3, ERBB4, PIK3CA, and PTEN) associated genes in patients with PeCa from a high incidence region in Brazil (Maranhão). HPV genotyping was performed by nest-PCR and genome sequencing, copy number alterations (CNAs) by array comparative genomic hybridization and gene copy number status, gene, and protein expression by quantitative polymerase chain reaction, reverse transcriptase-quantitative polymerase chain reaction, and immunohistochemistry, respectively. HPV genotyping revealed one of the highest frequencies of HPV reported in PeCa, affecting 96.4% of the cases. The most common CNAs observed were located at the HPV integration sites, such as 2p12-p11.2 and 14q32.33, where ADAM 6, KIAA0125, LINC00226, LINC00221, and miR7641-2, are mapped. Increased copy number of ERBB3 and EGFR genes were observed in association with COX2 and EGFR overexpression, reinforcing the role of the inflammatory pathway in PeCa, and suggesting anti-EGFR and anti-COX2 inhibitors as promising therapies for PeCa. Additionally, TP53 and RB1 messenger RNA downregulation was observed, suggesting the occurrence of other mechanisms for repression of these oncoproteins, in addition to the canonical HPV/TP53/RB1 signaling pathway. Our data reinforce the role of epigenetic events in abnormal gene expression in HPV-associated carcinomas and suggest the pivotal role of HPV driving CNAs and controlling gene expression in PeCa.

DeepMF: deciphering the latent patterns in omics profiles with a deep learning method

BACKGROUND

With recent advances in high-throughput technologies, matrix factorization techniques are increasingly being utilized for mapping quantitative omics profiling matrix data into low-dimensional embedding space, in the hope of uncovering insights in the underlying biological processes. Nevertheless, current matrix factorization tools fall short in handling noisy data and missing entries, both deficiencies that are often found in real-life data.

RESULTS

Here, we propose DeepMF, a deep neural network-based factorization model. DeepMF disentangles the association between molecular feature-associated and sample-associated latent matrices, and is tolerant to noisy and missing values. It exhibited feasible cancer subtype discovery efficacy on mRNA, miRNA, and protein profiles of medulloblastoma cancer, leukemia cancer, breast cancer, and small-blue-round-cell cancer, achieving the highest clustering accuracy of 76%, 100%, 92%, and 100% respectively. When analyzing data sets with 70% missing entries, DeepMF gave the best recovery capacity with silhouette values of 0.47, 0.6, 0.28, and 0.44, outperforming other state-of-the-art MF tools on the cancer data sets Medulloblastoma, Leukemia, TCGA BRCA, and SRBCT. Its embedding strength as measured by clustering accuracy is 88%, 100%, 84%, and 96% on these data sets, which improves on the current best methods 76%, 100%, 78%, and 87%.

CONCLUSION

DeepMF demonstrated robust denoising, imputation, and embedding ability. It offers insights to uncover the underlying biological processes such as cancer subtype discovery. Our implementation of DeepMF can be found at https://github.com/paprikachan/DeepMF.