The construction of a novel prognostic prediction model for glioma based on GWAS-identified prognostic-related risk loci

Backgrounds

Glioma is a highly malignant brain tumor with a grim prognosis. Genetic factors play a role in glioma development. While some susceptibility loci associated with glioma have been identified, the risk loci associated with prognosis have received less attention. This study aims to identify risk loci associated with glioma prognosis and establish a prognostic prediction model for glioma patients in the Chinese Han population.

Methods

A genome-wide association study (GWAS) was conducted to identify risk loci in 484 adult patients with glioma. Cox regression analysis was performed to assess the association between GWAS-risk loci and overall survival as well as progression-free survival in glioma. The prognostic model was constructed using LASSO Cox regression analysis and multivariate Cox regression analysis. The nomogram model was constructed based on the single nucleotide polymorphism (SNP) classifier and clinical indicators, enabling the prediction of survival rates at 1-year, 2-year, and 3-year intervals. Additionally, the receiver operator characteristic (ROC) curve was employed to evaluate the prediction value of the nomogram. Finally, functional enrichment and tumor-infiltrating immune analyses were conducted to examine the biological functions of the associated genes.

Results

Our study found suggestive evidence that a total of 57 SNPs were correlated with glioma prognosis (p < 5 × 10-5). Subsequently, we identified 25 SNPs with the most significant impact on glioma prognosis and developed a prognostic model based on these SNPs. The 25 SNP-based classifier and clinical factors (including age, gender, surgery, and chemotherapy) were identified as independent prognostic risk factors. Subsequently, we constructed a prognostic nomogram based on independent prognostic factors to predict individualized survival. ROC analyses further showed that the prediction accuracy of the nomogram (AUC = 0.956) comprising the 25 SNP-based classifier and clinical factors was significantly superior to that of each individual variable.

Conclusion

We identified a SNP classifier and clinical indicators that can predict the prognosis of glioma patients and established a prognostic prediction model in the Chinese Han population. This study offers valuable insights for clinical practice, enabling improved evaluation of patients' prognosis and informing treatment options.

Transcriptomic Profiling Reveals an Enhancer RNA Signature for Recurrence Prediction in Colorectal Cancer

BACKGROUND

Colorectal cancer (CRC) is one of the most fatal malignancies worldwide, and this is in part due to high rates of tumor recurrence in these patients. Currently, TNM staging remains the gold standard for predicting prognosis and recurrence in CRC patients; however, this approach is inadequate for identifying high-risk patients with the highest likelihood of disease recurrence. Recent evidence has revealed that enhancer RNAs (eRNAs) represent a higher level of cellular regulation, and their expression is frequently dysregulated in several cancers, including CRC. However, the clinical significance of eRNAs as recurrence predictor biomarkers in CRC remains unexplored, which is the primary aim of this study.

RESULTS

We performed a systematic analysis of eRNA expression profiles in colon cancer (CC) and rectal cancer (RC) patients from the TCGA dataset. By using rigorous biomarker discovery approaches by splitting the entire dataset into a training and testing cohort, we identified a 22-eRNA panel in CC and a 19-eRNA panel in RC for predicting tumor recurrence. The Kaplan-Meier analysis showed that biomarker panels robustly stratified low and high-risk CC (p = 7.29 × 10-5) and RC (p = 6.81 × 10-3) patients with recurrence. Multivariate and LASSO Cox regression models indicated that both biomarker panels were independent predictors of recurrence and significantly superior to TNM staging in CC (HR = 11.89, p = 9.54 × 10-4) and RC (HR = 3.91, p = 3.52 × 10-2). Notably, the ROC curves demonstrated that both panels exhibited excellent recurrence prediction accuracy in CC (AUC = 0.833; 95% CI: 0.74-0.93) and RC (AUC = 0.834; 95% CI: 0.72-0.92) patients. Subsequently, a combination signature that included the eRNA panels and TNM staging achieved an even greater predictive accuracy in patients with CC (AUC = 0.85).

CONCLUSIONS

Herein, we report a novel eRNA signature for predicting recurrence in patients with CRC. Further experimental validation in independent clinical cohorts, these biomarkers can potentially improve current risk stratification approaches for guiding precision oncology treatments in patients suffering from this lethal malignancy.

Alternative splicing patterns reveal prognostic indicator in muscle-invasive bladder cancer

Background

Bladder cancer is one of the most lethal malignancy in urological system, and 20–25% of bladder cancer patients are muscle invasive with unfavorable prognosis. However, the role of alternative splicing (AS) in muscle-invasive bladder cancer (MIBC) remains to be elucidated.

Methods

Percent spliced in (PSI) data obtained from the Cancer Genome Atlas (TCGA) SpliceSeq database (n = 394) were utilized to evaluate the AS events in MIBC. Prognosis-associated AS events were screened out by univariate Cox regression. LASSO Cox regression was used to identify reliable prognostic patterns in a training set and further validated in a test set. Splicing regulatory networks were constructed by correlations between PSI of AS events and RNA expression of splicing factors.

Results

As a result, a total of 2589 prognosis-related AS events in MIBC were identified. Pathways of spliceosomal complex (FDR = 0.017), DNA-directed RNA polymerase II, core complex (FDR = 0.032), and base excision repair (FDR = 0.038) were observed to be significantly enriched. Additionally, we noticed that most of the prognosis-related AS events were favorable factors. According to the LASSO and multivariate Cox regression analyses, 15-AS-based signature was established with the area under curve (AUC) of 0.709, 0.823, and 0.857 at 1-, 3-, and 5- years, respectively. The MIBC patients were further divided into high- and low-risk groups based on median risk sores. Interestingly, we observed that the prevalence of FGFR3 with mutations and focal amplification was significantly higher in low-risk group. Functional and immune infiltration analysis suggested potential signaling pathways and distinct immune states between these two groups. Moreover, splicing correlation network displayed a regulatory mode of prognostic splicing factors (SF) in MIBC patients.

Conclusions

This study not only provided novel insights into deciphering the possible mechanism of tumorgenesis and pathogenesis but also help refine risk stratification systems and potential treatment of decision-making for MIBC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12957-022-02685-0.

A Predictor Combining Clinical and Genetic Factors for AML1-ETO Leukemia Patients

Core Binding Factor (CBF)-AML is one of the most common somatic mutations in acute myeloid leukemia (AML). t(8;21)/AML1-ETO-positive acute myeloid leukemia accounts for 5-10% of all AMLs. In this study, we consecutively included 254 AML1-ETO patients diagnosed and treated at our institute from December 2009 to March 2020, and evaluated molecular mutations by 185-gene NGS platform to explore genetic co-occurrences with clinical outcomes. Our results showed that high KIT VAF(≥15%) correlated with shortened overall survival compared to other cases with no KIT mutation (3-year OS rate 26.6% vs 59.0% vs 69.6%, HR 1.50, 95%CI 0.78-2.89, P=0.0005). However, no difference was found in patients’ OS whether they have KIT mutation in two or three sites. Additionally, we constructed a risk model by combining clinical and molecular factors; this model was validated in other independent cohorts. In summary, our study showed that c-kit other than any other mutations would influence the OS in AML1-ETO patients. A proposed predictor combining both clinical and genetic factors is applicable to prognostic prediction in AML1-ETO patients.

Identification and Prognostic Value Exploration of Cyclophosphamide (Cytoxan)-Centered Chemotherapy Response-Associated Genes in Breast Cancer

In this study, we aimed to explore cyclophosphamide (Cytoxan) response-associated genes and constructed a model to predict the prognosis of breast cancer (BRCA) patients. Samples obtained from TCGA and GEO databases were subjected to Weighted Gene Coexpression Network Analysis (WGCNA) and univariate Cox and LASSO Cox regression analysis to identify and validate the Cytoxan response-related prognostic signature. Moreover, multivariate Cox regression analysis was performed to analyze the independence of factors, and the nomogram model was constructed by including all the independent factors. WGCNA revealed that 159 genes are significantly correlated with Cytoxan response in BRCA samples, and the samples with a different prognosis could be effectively distinguished based on the expression of those 159 genes. Ten genes were further selected to be related to the prognosis of BRCA patients, including PCDHB2, GRIK2, FRMD7, CCSER1, PCDHGA1, PCDHA1, LRRC37A6P, PCDHGA12, ZNF486, and PCDHGB5, based on the Risk Score model. Among them, PCDHA1 expression was validated in cells and patient samples. Multivariate Cox regression analysis confirmed that the Risk Score is an independent factor. Furthermore, the nomogram model showed that the predicted survival probability is closely related to the actual survival probability. In conclusion, we identified 159 genes potentially correlated with the Cytoxan response of BRCA patients, which had prognostic value in BRCA.

Immune-Related Genes Are Prognostic Markers for Prostate Cancer Recurrence

Background

Prostate cancer (PCa) is an immune-responsive disease. The current study sought to explore a robust immune-related prognostic gene signature for PCa.

Methods

Data were retrieved from the tumor Genome Atlas (TCGA) database and GSE46602 database for performing the least absolute shrinkage and selection operator (LASSO) cox regression model analysis. Immune related genes (IRGs) data were retrieved from ImmPort database.

Results

The weighted gene co-expression network analysis (WGCNA) showed that nine functional modules are correlated with the biochemical recurrence of PCa, including 259 IRGs. Univariate regression analysis and survival analysis identified 35 IRGs correlated with the prognosis of PCa. LASSO Cox regression model analysis was used to construct a risk prognosis model comprising 18 IRGs. Multivariate regression analysis showed that risk score was an independent predictor of the prognosis of PCa. A nomogram comprising a combination of this model and other clinical features showed good prediction accuracy in predicting the prognosis of PCa. Further analysis showed that different risk groups harbored different gene mutations, differential transcriptome expression and different immune infiltration levels. Patients in the high-risk group exhibited more gene mutations compared with those in the low-risk group. Patients in the high-risk groups showed high-frequency mutations in TP53. Immune infiltration analysis showed that M2 macrophages were significantly enriched in the high-risk group implying that it affected prognosis of PCa patients. In addition, immunostimulatory genes were differentially expressed in the high-risk group compared with the low-risk group. BIRC5, as an immune-related gene in the prediction model, was up-regulated in 87.5% of prostate cancer tissues. Knockdown of BIRC5 can inhibit cell proliferation and migration.

Conclusion

In summary, a risk prognosis model based on IGRs was developed. A nomogram comprising a combination of this model and other clinical features showed good accuracy in predicting the prognosis of PCa. This model provides a basis for personalized treatment of PCa and can help clinicians in making effective treatment decisions.

Exploration of the Role of m6 A RNA Methylation Regulators in Malignant Progression and Clinical Prognosis of Ovarian Cancer

Ovarian cancer is the most deadly gynecologic malignancy worldwide and it is warranted to dissect the critical gene regulatory network in ovarian cancer. N6-methyladenosine (m⁶A) RNA methylation, as the most prevalent RNA modification, is orchestrated by the m⁶A RNA methylation regulators and has been implicated in malignant progression of various cancers. In this study, we investigated the genetic landscape and expression profile of the m⁶A RNA methylation regulators in ovarian cancer and found that several m⁶A RNA methylation regulators were frequently amplified and up-regulated in ovarian cancer. Utilizing consensus cluster analysis, we stratified ovarian cancer samples into four clusters with distinct m⁶A methylation patterns and patients in these subgroups displayed the different clinical outcomes. Moreover, multivariate Cox proportional hazard model was used to screen the key m⁶A regulators associated with the prognosis of ovarian cancer and the last absolute shrinkage and selection operator (LASSO) Cox regression was used to construct the gene signature for prognosis prediction. The survival analysis exhibited the risk-gene signature could be used as independent prognostic markers for ovarian cancer. In conclusion, m⁶A RNA methylation regulators are associated with the malignant progression of ovarian cancer and could be a potential in prognostic prediction for ovarian cancer.

More evidence for prediction model of radiosensitivity

With the development of precision medicine, searching for potential biomarkers plays a major role in personalized medicine. Therefore, how to predict radiosensitivity to improve radiotherapy is a burning question. The definition of radiosensitivity is complex. Radiosensitive gene/biomarker can be useful for predicting which patients would benefit from radiotherapy. The discovery of radiosensitivity biomarkers require multiple pieces of evidence. A prediction model of breast cancer radiosensitivity based on six genes was established. We had put forward some supplements on the basis of the present study. We found that there were no differences between high- and low-risk scores in the non-radiotherapy group. Patients who received radiotherapy had a significantly better overall survival than non-radiotherapy patients in the predicted low-risk score patients. Furthermore, there was no difference between radiotherapy group and non-radiotherapy group in the high-risk score group. Those results firmly supported the prediction model of radiosensitivity. In addition, building a radiosensitivity prediction model was systematically discussed. Genes of model could be screened by different methods, such as Cox regression analysis, Lasso Cox regression method, random forest algorithm and other methods. In the future, precision radiotherapy might depend on the combination of multi-omics data and high dimensional image data.

Identification of a novel autophagy signature for predicting survival in patients with lung adenocarcinoma

Background

Lung adenocarcinoma (LUAD) is the most commonhistological lung cancer subtype, with an overall five-year survivalrate of only 17%. In this study, we aimed to identify autophagy-related genes (ARGs) and develop an LUAD prognostic signature.

Methods

In this study, we obtained ARGs from three databases and downloaded gene expression profiles from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) database. We used TCGA-LUAD (n = 490) for a training and testing dataset, and GSE50081 (n = 127) as the external validation dataset.The least absolute shrinkage and selection operator (LASSO) Cox and multivariate Cox regression models were used to generate an autophagy-related signature. We performed gene set enrichment analysis (GSEA) and immune cell analysis between the high- and low-risk groups. A nomogram was built to guide the individual treatment for LUAD patients.

Results

We identified a total of 83 differentially expressed ARGs (DEARGs) from the TCGA-LUAD dataset, including 33 upregulated DEARGs and 50 downregulated DEARGs, both with thresholds of adjusted P < 0.05 and |Fold change| > 1.5. Using LASSO and multivariate Cox regression analyses, we identified 10 ARGs that we used to build a prognostic signature with areas under the curve (AUCs) of 0.705, 0.715, and 0.778 at 1, 3, and 5 years, respectively. Using the risk score formula, the LUAD patients were divided into low- or high-risk groups. Our GSEA results suggested that the low-risk group were enriched in metabolism and immune-related pathways, while the high-risk group was involved in tumorigenesis and tumor progression pathways. Immune cell analysis revealed that, when compared to the high-risk group, the low-risk group had a lower cell fraction of M0- and M1- macrophages, and higher CD4 and PD-L1 expression levels.

Conclusion

Our identified robust signature may provide novel insight into underlying autophagy mechanisms as well as therapeutic strategies for LUAD treatment.

A six-gene-based signature for breast cancer radiotherapy sensitivity estimation

Breast cancer (BRCA) represents the most common malignancy among women worldwide with high mortality. Radiotherapy is a prevalent therapeutic for BRCA that with heterogeneous effectiveness among patients. Here, we proposed to develop a gene expression-based signature for BRCA radiotherapy sensitivity estimation. Gene expression profiles of BRCA samples from the Cancer Genome Atlas (TCGA) and International Cancer Genome Consortium (ICGC) were obtained and used as training and independent testing dataset, respectively. Differential expression genes (DEGs) in BRCA samples compared with their paracancerous samples in the training set were identified by using the edgeR Bioconductor package. Univariate Cox regression analysis and LASSO Cox regression method were applied to screen optimal genes for constructing a radiotherapy sensitivity estimation signature. Nomogram combining independent prognostic factors was used to predict 1-, 3-, and 5-year OS of radiation-treated BRCA patients. Relative proportions of tumor infiltrating immune cells (TIICs) calculated by CIBERSORT and mRNA levels of key immune checkpoint receptors was adopted to explore the relation between the signature and tumor immune response. As a result, 603 DEGs were obtained in BRCA tumor samples, six of which were retained and used to construct the radiotherapy sensitivity prediction model. The signature was proved to be robust in both training and testing sets. In addition, the signature was closely related to the immune microenvironment of BRCA in the context of TIICs and immune checkpoint receptors’ mRNA levels. In conclusion, the present study obtained a radiotherapy sensitivity estimation signature for BRCA, which should shed new light in clinical and experimental research.

The Identification of Critical m6A RNA Methylation Regulators as Malignant Prognosis Factors in Prostate Adenocarcinoma

RNA methylation accounts for over 60% of all RNA modifications, and N⁶-methyladenosine (m⁶A) is the most common modification on mRNA and lncRNA of human beings. It has been found that m⁶A modification occurs in microRNA, circRNA, rRNA, and tRNA, etc. The m⁶A modification plays an important role in regulating gene expression, and the abnormality of its regulatory mechanism refers to many human diseases, including cancers. Pitifully, as it stands there is a serious lack of knowledge of the extent to which the expression and function of m⁶A RNA methylation can influence prostate cancer (PC). Herein, we systematically analyzed the expression levels of 35 m⁶A RNA methylation regulators mentioned in literatures among prostate adenocarcinoma patients in the Cancer Genome Atlas (TCGA), finding that most of them expressed differently between cancer tissues and normal tissues with the significance of p < 0.05. Utilizing consensus clustering, we divided PC patients into two subgroups based on the differentially expressed m⁶A RNA methylation regulators with significantly different clinical outcomes. To appraise the discrepancy in total transcriptome between subgroups, the functional enrichment analysis was conducted for differential signaling pathways and cellular processes. Next, we selected five critical genes by the criteria that the regulators had a significant impact on prognosis of PC patients from TCGA through the last absolute shrinkage and selection operator (LASSO) Cox regression and obtained a risk score by weighted summation for prognosis prediction. The survival analysis curve and receiver operating characteristic (ROC) curve showed that this signature could excellently predict the prognosis of PC patients. The univariate and multivariate Cox regression analyses proved the independent prognostic value of the signature. In summary, our effort revealed the significance of m⁶A RNA methylation regulators in prostate cancer and determined a m⁶A gene expression classifier that well predicted the prognosis of prostate cancer.

Gene signature for prognosis in comparison of pancreatic cancer patients with diabetes and non-diabetes

BACKGROUND

Pancreatic cancer (PC) has much weaker prognosis, which can be divided into diabetes and non-diabetes. PC patients with diabetes mellitus will have more opportunities for physical examination due to diabetes, while pancreatic cancer patients without diabetes tend to have higher risk. Identification of prognostic markers for diabetic and non-diabetic pancreatic cancer can improve the prognosis of patients with both types of pancreatic cancer.

METHODS

Both types of PC patients perform differently at the clinical and molecular levels. The Cancer Genome Atlas (TCGA) is employed in this study. The gene expression of the PC with diabetes and non-diabetes is used for predicting their prognosis by LASSO (Least Absolute Shrinkage and Selection Operator) Cox regression. Furthermore, the results are validated by exchanging gene biomarker with each other and verified by the independent Gene Expression Omnibus (GEO) and the International Cancer Genome Consortium (ICGC). The prognostic index (PI) is generated by a combination of genetic biomarkers that are used to rank the patient's risk ratio. Survival analysis is applied to test significant difference between high-risk group and low-risk group.

RESULTS

An integrated gene prognostic biomarker consisted by 14 low-risk genes and six high-risk genes in PC with non-diabetes. Meanwhile, and another integrated gene prognostic biomarker consisted by five low-risk genes and three high-risk genes in PC with diabetes. Therefore, the prognostic value of gene biomarker in PC with non-diabetes and diabetes are all greater than clinical traits (HR = 1.102, P-value < 0.0001; HR = 1.212, P-value < 0.0001). Gene signature in PC with non-diabetes was validated in two independent datasets.

CONCLUSIONS

The conclusion of this study indicated that the prognostic value of genetic biomarkers in PCs with non-diabetes and diabetes. The gene signature was validated in two independent databases. Therefore, this study is expected to provide a novel gene biomarker for predicting prognosis of PC with non-diabetes and diabetes and improving clinical decision.

A prognostic eight-gene expression signature for patients with breast cancer receiving adjuvant chemotherapy

Breast cancer is a popularly diagnosed malignant tumor. Genomic profiling studies suggest that breast cancer is a disease with heterogeneity. Chemotherapy is one of the chief means to treat breast cancer, while its responses and clinical outcomes vary largely due to the conventional clinicopathological factors and inherent chemosensitivity of breast cancer. Using the least absolute shrinkage and selection operator (LASSO) Cox regression model, our study established a multi-mRNA-based signature model and constructed a relative nomogram in predicting distant-recurrence-free survival for patients receiving surgery and following chemotherapy. We constructed a signature of eight mRNAs (IPCEF1, SYNDIG1, TIGIT, SPESP1, C2CD4A, CLCA2, RLN2, and CCL19) with the LASSO model, which was employed to separate subjects into groups with high- and low-risk scores. Obvious differences of distant-recurrence-free survival were found between these two groups. This eight-mRNA-based signature was independently associated with the prognosis and had better prognostic value than classical clinicopathologic factors according to multivariate Cox regression results. Receiver operating characteristic results demonstrated excellent performance in diagnosing 3-year distant-recurrence by the eight-mRNA signature. A nomogram that combined both the eight-mRNA-based signature and clinicopathological risk factors was constructed. Comparing with an ideal model, the nomograms worked well both in the training and validation sets. Through the results that the eight-mRNA signature effectively classified patients into low- and high-risk of distant recurrence, we concluded that this eight-mRNA-based signature played a promising predictive role in prognosis and could be clinically applied in breast cancer patients receiving adjuvant chemotherapy.

A Gene Signature of Survival Prediction for Kidney Renal Cell Carcinoma by Multi-Omic Data Analysis

Kidney renal cell carcinoma (KIRC), which is the most common subtype of kidney cancer, has a poor prognosis and a high mortality rate. In this study, a multi-omics analysis is performed to build a multi-gene prognosis signature for KIRC. A combination of a DNA methylation analysis and a gene expression data analysis revealed 863 methylated differentially expressed genes (MDEGs). Seven MDEGs (BID, CCNF, DLX4, FAM72D, PYCR1, RUNX1, and TRIP13) were further screened using LASSO Cox regression and integrated into a prognostic risk score model. Then, KIRC patients were divided into high- and low-risk groups. A univariate cox regression analysis revealed a significant association between the high-risk group and a poor prognosis. The time-dependent receiver operating characteristic (ROC) curve shows that the risk group performs well in predicting overall survival. Furthermore, the risk group is contained in the best multivariate model that was obtained by a multivariate stepwise analysis, which further confirms that the risk group can be used as a potential prognostic biomarker. In addition, a nomogram was established for the best multivariate model and shown to perform well in predicting the survival of KIRC patients. In summary, a seven-MDEG signature is a powerful prognosis factor for KIRC patients and may provide useful suggestions for their personalized therapy.

Co-Expression Network Analysis Identified Gene Signatures in Osteosarcoma as a Predictive Tool for Lung Metastasis and Survival

Osteosarcoma (OS) is the most common primary bone tumor, whose poor prognosis is mainly due to lung metastasis. The aim of this study is to build a practical and valid diagnostic test that can predict the risk of OS metastasis and progression. We performed weighted gene co-expression network analysis (WGCNA) on GSE21257 from the Gene Expression Omnibus (GEO) database, which contains microarray data of biopsies from OS patients. In these modules, the highest association was found between the blue module and metastasis stage (r = -0.52) by Pearson's correlation analysis. Based on Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression, we derived eight clinically significant genes and constructed an eight-gene signature for metastasis status. It showed great efficacy to distinguish metastasis from non-metastasis (AUC = 0.886) and the results were validated in The Cancer Genome Atlas (TCGA) database. Functional enrichment analysis of hub genes showed that their biological processes focused on immune-related pathways, suggesting the important roles of immune cells, immune pathways and the tumor microenvironment in metastasis development. In conclusion, we discovered an efficient gene signature with great efficacy to distinguish metastasis status, which may help improve early diagnosis and treatment, enhancing the clinical outcomes of OS patients. Besides we created an effective protocol to seek for several hub genes in high-throughput data by combining WGCNA and LASSO Cox regression.

Prognostic and Predictive Value of Three DNA Methylation Signatures in Lung Adenocarcinoma

Background: Lung adenocarcinoma (LUAD) is the leading cause of cancer-related mortality worldwide. Molecular characterization-based methods hold great promise for improving the diagnostic accuracy and for predicting treatment response. The DNA methylation patterns of LUAD display a great potential as a specific biomarker that will complement invasive biopsy, thus improving early detection. Method: In this study, based on the whole-genome methylation datasets from The Cancer Genome Atlas (TCGA) and several machine learning methods, we evaluated the possibility of DNA methylation signatures for identifying lymph node metastasis of LUAD, differentiating between tumor tissue and normal tissue, and predicting the overall survival (OS) of LUAD patients. Using the regularized logistic regression, we built a classifier based on the 3616 CpG sites to identify the lymph node metastasis of LUAD. Furthermore, a classifier based on 14 CpG sites was established to differentiate between tumor and normal tissues. Using the Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression, we built a 16-CpG-based model to predict the OS of LUAD patients. Results: With the aid of 3616-CpG-based classifier, we were able to identify the lymph node metastatic status of patients directly by the methylation signature from the primary tumor tissues. The 14-CpG-based classifier could differentiate between tumor and normal tissues. The area under the receiver operating characteristic (ROC) curve (AUC) for both classifiers achieved values close to 1, demonstrating the robust classifier effect. The 16-CpG-based model showed independent prognostic value in LUAD patients. Interpretation: These findings will not only facilitate future treatment decisions based on the DNA methylation signatures but also enable additional investigations into the utilization of LUAD DNA methylation pattern by different machine learning methods.