Comprehensive pan-cancer multi-omics analysis of ATAD2 in human cancers

Abstract

Background:

ATAD2, a member of the AAA + ATPase family and a known cancer/testicular antigen, has demonstrated oncogenic properties in various cancers. Despite its established role in cancer progression, a comprehensive pan-cancer analysis of its prognostic significance, therapeutic predictive potential, and immunological associations has not been conducted.

Methods:

We performed a pan-cancer analysis of ATAD2 using data from The Cancer Genome Atlas (TCGA), assessing its expression profiles across different tumor types. We examined its correlations with survival prognosis, clinical outcomes, and cancer-associated signaling pathways. Additionally, we explored the relationship between ATAD2 expression and the tumor immune microenvironment, as well as its predictive value for responses to immunotherapy, targeted therapy, and chemotherapy.

Results:

ATAD2 was significantly upregulated in multiple tumor types, and its increased expression was associated with poor prognosis and adverse clinical outcomes. Further analysis revealed that ATAD2 expression was closely linked to cancer-related signaling pathways, tumor mutational burden, and microsatellite instability. Moreover, ATAD2 expression showed strong correlations with immune cell infiltration, immune-related gene expression, and responses to various treatment regimens, including immunotherapy, targeted therapy, and chemotherapy.

Conclusion:

ATAD2 is a critical biomarker for poor prognosis and treatment response across a wide range of cancers. Its ability to predict therapeutic responses, particularly for immunotherapy, targeted therapy, and chemotherapy, underscores its clinical significance. Further experimental studies and clinical trials are needed to explore ATAD2’s role in cancer progression and its involvement in therapeutic resistance mechanisms.

1. Introduction

Cancer poses a significant threat to human health worldwide. According to the World Health Organization, 2020 saw an alarming 19.3 million new cancer diagnoses. Tragically, this devastating illness resulted in nearly 10 million deaths in 2019, making it second only to cardiovascular disease as the leading cause of mortality.^[1] Treatment tolerance and disease recurrence further increase the risk of mortality for cancer patients.^[2] Traditional cancer therapies, including surgery, radiation therapy, and chemotherapy, face limitations despite their efficacy. These treatments can induce significant side effects and demonstrate reduced effectiveness in patients with advanced disease. Immunotherapy, particularly using immune checkpoint inhibitors (ICBs) and chimeric antigen receptor T-cell (CART) therapy, has achieved remarkable breakthroughs in cancer treatment in recent years.^[3] However, some patients still experience limited or no response to immunotherapy.^[4] Therefore, identifying novel immune biomarkers is crucial to improving the safety and effectiveness of cancer immunotherapy. Pan-cancer analysis is a broad approach that combines different types of cancer data for a comprehensive view. Through multi-omics analysis, we aim to highlight the common characteristics of cancer, focusing on its universal features and distinctions.

ATPase Family AAA Domain-Containing Protein 2 (ATAD2), a member of the AAA + ATPase family, is considered a cancer/testis antigen.^[5] ATAD2 is implicated in the orchestration of cell cycle dynamics, transcriptional regulation, and DNA replication process.^[6,7] Through its bromine-binding domain, ATAD2 is capable of recognizing acetylated histones, thereby participating in chromatin remodeling and the regulation of gene expression.^[7] ATAD2 can modulate the activity of specific transcription factors, resulting in the regulation of oncogene expression.^[8] Research has shown that ATAD2 expression is markedly increased in a range of malignancies, including ovarian cancer, pancreatic cancer, esophageal squamous cell carcinoma, colorectal cancer (CRC), breast cancer, and lung adenocarcinoma (LUAD).^[9–14] Despite growing interest in ATAD2’s expression and function in cancer, its precise mechanisms in tumor development remain poorly understood, particularly at the pan-cancer level. In this study, we significantly expand upon previous findings by leveraging a more comprehensive dataset and employing a multi-omics approach to explore ATAD2’s role across multiple cancer types.^[15] Our analysis delves into previously underexplored areas, including locus-specific DNA methylation, therapeutic response predictions across immunotherapy, targeted therapy, and chemotherapy, and the intricate relationships between ATAD2 expression, immune infiltration, and tumor microenvironment dynamics. By integrating diverse analytical dimensions, our research provides a systematic and in-depth perspective on ATAD2’s involvement in cancer progression and therapeutic resistance, offering a solid foundation for future experimental and translational research.

In conducting this comprehensive pan-cancer analysis, we employed a multi-omics-based approach to functionally characterize ATAD2. Our study encompassed key aspects such as gene expression profiling, survival prognosis, gene enrichment analysis, genetic alterations, DNA methylation patterns, immune infiltration, and therapeutic response predictions for immunotherapy, targeted therapy, and chemotherapy. This holistic framework enables us to uncover novel insights into ATAD2’s multifaceted role in cancer biology.

2. Methods

2.1. Dataset collection and gene expression analysis

RNA sequencing and phenotypic profiles were obtained from the UCSC Xena browser (https://xenabrowser.net/), which integrates data from The Cancer Genome Atlas (TCGA)^[16] and the Genotype-Tissue Expression (GTEx)^[17] database, effectively removing batch effects.^[18] Data from the Cancer Cell Line Encyclopedia (CCLE)^[19] were downloaded from the DepMap Portal (https://depmap.org/portal/). Immunofluorescence images were obtained from the Human Protein Atlas (HPA, https://www.proteinatlas.org/)^[20] database. Furthermore, we utilized the ProteoCancer Analysis Suite (PCAS)^[21] to analyze proteomic data for multiple tumor types from the Clinical Proteomic Tumor Analysis Consortium (CPTAC, https://proteomics.cancer.gov/programs/cptac)^[22] database. Additionally, external validation was conducted using 7 publicly available gene expression datasets the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) database, including the glioblastoma multiforme (GBM) dataset GSE7696, the head and neck squamous cell carcinoma (HNSC) dataset GSE12452, the uterine corpus endometrial carcinoma (UCEC) dataset GSE17025, the CRC dataset GSE37182, the LUAD dataset GSE13213 and GSE40791, the breast invasive carcinoma (BRCA) dataset GSE42568, the liver hepatocellular carcinoma (LIHC) dataset GSE112790, and GSE76427, the lung squamous cell carcinoma (LUSC) dataset GSE19188, the cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC) dataset GSE63514, the adrenocortical carcinoma (ACC) dataset GSE10927, the kidney renal clear cell carcinoma (KIRC) dataset GSE167573, the ovarian serous cystadenocarcinoma (OV) dataset GSE12470, the esophageal carcinoma (ESCA) dataset GSE53624, the pancreatic adenocarcinoma (PAAD) dataset GSE183795, the sarcoma (SARC) dataset GSE30929, and the uveal melanoma (UVM) dataset GSE22138. Additionally, brain lower-grade glioma (LGG) dataset CGGA693 was obtained from the Chinese Glioma Genome Atlas (CGGA, http://www.cgga.org.cn/).^[23] The information of all datasets was shown in Table S1, Supplemental Digital Content, https://links.lww.com/MD/P455. These datasets were analyzed using the “limma” package to identify differentially expressed genes with a P-value < .05 and |Log2Fold change| > 1, indicating statistical significance.

2.2. Prognosis and diagnosis analysis

Survival data for pan-cancer, including overall survival (OS), disease-specific survival (DSS), disease-free interval (DFI), and progression-free interval (PFI), were obtained from the TCGA database. Cox analysis and Kaplan–Meier (KM) survival curves were generated using the R packages “survival” and “survminer” to evaluate the association between ATAD2 expression and patient prognosis. The “pROC” package in R produces a receiver operating characteristic (ROC) curve to evaluate the performance of a diagnostic test. A diagnostic test with an area under the curve value >0.7 is considered highly accurate and reliable.

2.3. Clinical features analysis

The correlation between ATAD2 mRNA expression and various clinicopathological features, including gender, clinical stage, and grade, was analyzed across multiple cancer types using the Sanger Box website (http://sangerbox.com/home.html).^[24]

2.4. Gene set enrichment analysis

To assess the biological function of ATAD2 in tumors, we performed gene set enrichment analysis (GSEA) using Kyoto Encyclopedia of Genes and Genomes (KEGG) and biological process gene sets (c2.cp.kegg_medicus.v2024.1.Hs.entrez and c5.go.bp.v2024.1.Hs.entrez, respectively) from the Molecular Signatures Database (MSigDB). The analysis was conducted in R using ggplot2, org.Hs.eg.db, clusterProfiler, and enrichplot for visualization and statistical analysis.

2.5. Gene alteration analysis

We sourced the copy number variation (CNV) and DNA methylation profile data from the UCSC Xena browser. Data on pan-cancer tumor mutation burden (TMB) and microsatellite instability (MSI) were obtained from the SangerBox platform. To investigate the associations between ATAD2 expression and TMB, MSI, DNA methylation, and CNV, we utilized the “ggplot2,” “fmsb,” and ‘biomaRt’ packages in R for analysis and visualization.

2.6. Immune infiltration analysis

The pan-cancer immune cell infiltration data were obtained from the Tumor Immune Estimation Resource 2.0 (TIMER2.0, http://timer.cistrome.org/).^[25] We analyzed the correlation between ATAD2 mRNA expression levels and tumor-infiltrating immune cell infiltration by xCell algorithm.^[26]

2.7. Immunological correlation analysis

To further investigate the associations between ATAD2 and immune-related gene expression, we conducted correlation analysis encompassing chemokine, chemokine receptors, immunoinhibitor, immunostimulator, and MHC genes. We obtained these gene sets from the TISIDB (http://cis.hku.hk/TISIDB/) database.^[27]

2.8. Therapeutic analysis of ATAD2

We collected chemotherapy-related data from the BRCA datasets GSE25055 and GSE25065, as well as the non-small cell lung cancer (NSCLC) dataset GSE42127. Additionally, data from the OV dataset GSE63885 focused on chemotherapy resistance. For hepatocellular carcinoma (LIHC), the dataset GSE104580 was used to analyze ATAD2 expression in patients undergoing transarterial chemoembolization (TACE), and the dataset GSE109211 provided information on sorafenib treatment responses. Furthermore, 2 immunotherapy cohorts were retrieved from supplementary files: one from the study by David (PMID:32472114)^[28] and another from the study by Robert (PMID:32895571),^[29] Supplemental Digital Content, https://links.lww.com/MD/P455.

2.9. Immunohistochemical staining

We investigated ATAD2 protein expression differences using the HPA database, analyzing staining levels in tumor and normal tissues from 12 organs: breast, cerebrum, colon, liver, lung, ovary, pancreas, endometrium, stomach, bladder, nasopharynx, and skin.

2.10. Statistical analysis

In this study, statistical analyses were carried out using R (version 4.2.3). Comparisons between groups were conducted using the Wilcoxon rank-sum test. Spearman rank correlation was applied for the correlation analysis, with |r| = 0.3 being deemed indicative of a significant correlation. A P-value of <0.05 was deemed statistically significant.

3. Result

3.1. Gene expression analysis in pan-cancer

To explore the expression of ATAD2, data from CCLE, GTEx, TCGA, and CPTAC were analyzed. The CCLE database analysis identified bone marrow, breast and lung neoplasms as the top 3 tumor cell lines with the highest ATAD2 expression (Fig. S1, Supplemental Digital Content, https://links.lww.com/MD/P456). In our analysis of normal tissues from GTEx, we observed obvious differences in ATAD2 RNA expression across various tissues, particularly noteworthy being the highest expression levels detected in bone marrow and testis, in contrast to the lowest levels found in kidney and blood (Fig. 1A). Furthermore, comparative analysis of ATAD2 mRNA levels between cancer and normal tissues was performed from TCGA database. The results showed that compared with normal tissues, ATAD2 mRNA expression was significantly upregulated in 16 cancer types (Fig. 1B). After combining the GTEx and TCGA samples, our findings revealed a significant upregulation of ATAD2 mRNA expression in 26 different types of cancer compared to normal tissues, including bladder urothelial carcinoma (BLCA), BRCA, CESC, cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), ESCA, HNSC, GBM, LGG, ACC, kidney renal papillary cell carcinoma (KIRP), KIRC, kidney chromophobe (KICH), LIHC, LUAD, LUSC, PAAD, Prostate Adenocarcinoma (PRAD), SARC, skin cutaneous melanoma (SKCM), thyroid carcinoma (THCA), rectal cancer (READ), stomach adenocarcinoma (STAD), OV, uterine carcinosarcoma (UCS) and UCEC (Fig. 1C). What’s more, we analyzed ATAD2 expression in tumors and their corresponding normal tissues. Remarkably, a consistent trend emerged, where 15 tumor tissues exhibited increased ATAD2 expression relative to their corresponding normal tissues (Fig. 1D). The consistently elevated expression of ATAD2 across multiple cancer types was further validated using GEO datasets (Fig. S2, Supplemental Digital Content, https://links.lww.com/MD/P456).

Figure 1.

ATAD2 mRNA expression differences. (A) ATAD mRNA expression in normal tissues was assessed via the GTEx database. (B) ATAD mRNA expression in pan-cancer was explored using TIMER2.0 based on the TCGA database (C) TCGA and GTEx databases were used to compare the expression differences of ATAD2 between tumor and normal tissues in pan-cancer. (D) Comparison of ATAD2 expression between paired tumor and normal samples from the TCGA database. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2, GTEx = Genotype-Tissue Expression, TCGA = The Cancer Genome Atlas.

Further analysis at the protein level, ATAT2 resides in the nucleoplasm of human cells (Fig. 2A). According to immunofluorescence assays, ATAD2 colocalizes with nuclear markers, suggesting its localization in the nucleoplasm of cancer cells (Fig. 2B). Additionally, ATAD2 protein expression was found to be upregulated in 14 datasets comprising 11 cancer types in the CPTAC database, including BRCA, KIRC, COAD, GBM, LIHC, HNSC, LUSC, LUAD, OV, PAAD and UCEC (Fig. 2C).

Figure 2.

ATAD2 protein localization and expression analysis. (A) ATAD2 protein’s subcellular positioning in the nucleoplasm. (B) Immunofluorescence reveals the nucleoplasm distribution of ATAD2 protein in cancer cells (C) ATAD protein expression in tumor and normal tissues across multiple cancers was analyzed using the CPTAC database. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2, CPTAC = Clinical Proteomic Tumor Analysis Consortium.

3.2. Prognosis and diagnosis analysis in pan-cancer

According to ROC curves, ATAD2 may serve as a useful diagnostic biomarker for certain cancers (Fig. S3, Supplemental Digital Content, https://links.lww.com/MD/P456). To explore the potential impact of ATAD2 expression on patient outcomes in different cancer types, we carried out a comprehensive survival analysis, evaluating OS, DSS, DFI, and PFI. Cox proportional hazards model analysis revealed a significant association between ATAD2 expression levels and OS in several cancer types: ACC, KICH, KIRP, LGG, LUAD, mesothelioma (MESO), PAAD and UVM (P < .05) (Fig. 3A). Notably, higher ATAD2 expression lever was identified as a risk factor for OS in these cancer types. Next, ATAD2 expression levels were significantly correlated with DSS in ACC, KICH, KIRP, LGG, LUAD, MESO, PAAD, PRAD and UVM (P < .05) (Fig. 3B). Then, we found a marked correlation between ATAD2 expression and DFI in CESC, KIRP, PAAD, SARC, and UVM (Fig. 3C). Lastly, ATAD2 expression correlated with PFI in several cancers, acting as a protective factor in GBM but as a risk factor in others, including ACC, KICH, KIRP, LGG, LIHC, MESO, PAAD, SARC, and UVM (Fig. 3D). KM curves were employed to evaluate the 4 prognostic outcomes. Our analysis of ACC, KIRP, LUAD, PAAD, and SARC revealed that patients with high ATAD2 expression had a significantly worse prognosis in terms of OS, DSS, PFI, and DFI. In KICH, LGG, LIHC, MESO, UCEC, and UVM, a high level of ATAD2 expression was associated with poor prognosis in at least 3 of the survival outcomes (Figs. S4–S7, Supplemental Digital Content, https://links.lww.com/MD/P456). Consistent findings were also obtained from the KM analysis across 7 datasets: GSE13213, GSE183795, CGGA693, GSE10927, GSE30929, GSE76427, and GSE22138 (Fig. 3E–G, Fig. S8A–D, Supplemental Digital Content, https://links.lww.com/MD/P456).

Figure 3.

The predictive significance of ATAD2 in pan-cancer patients. (A) Forest Plots for TCGA OS Outcomes. (B) Forest Plots for TCGA DSS Outcomes. (C) Forest Plots for TCGA PFI Outcomes. (D) Forest Plots for TCGA DFI Outcomes. Blue dots indicate protective factors, red dots signify risk factors, and gray dots denote factors with no significant effect. (E) Kaplan–Meier survival analysis for LUAD in GSE13213. (F) Kaplan–Meier survival analysis for PAAD in GSE183795. (G) Kaplan–Meier survival analysis for LGG in CGGA693. ATAD2 = ATPase Family AAA Domain Containing 2, DFI = disease-free interval, DSS = disease-specific survival, LGG = brain lower-grade glioma, LUAD = lung adenocarcinoma, OS = overall survival, PAAD = pancreatic adenocarcinoma, PFI = progression-free interval, TCGA = The Cancer Genome Atlas.

3.3. Clinical features analysis in pan-cancer

To determine if there is a link between ATAD2 expression and gender among pan-cancer patients, we assessed its expression in various cancer types across both genders. Studies indicate that ATAD2 expression levels are higher in male patients compared to females in cancers such as HNSC, LUAD, acute myeloid leukemia (LAML), and BRCA; conversely, ATAD2 expression is lower in males than in females in SARC, KIRP, and pan-kidney cohort (KIPAN) (P < .05) (Fig. 4A). In addition, to gain deeper insights into the relationship between ATAD2 and cancer progression, we conducted an analysis of ATAD2 expression across various types and stages of cancer, aiming to identify correlations with tumor stage and grade. Our findings identified significant variations in the expression of ATAD2 among 9 different tumor types, notably in LUAD, KIRP, KIRC, LUSC, testicular germ cell tumors (TGCT), ACC, UCS, and KICH (P < .05) (Fig. 4B). Grading analysis revealed observable differences across 9 tumor types, including glioma (GBMLGG), LGG, STES, STAD, UCEC, HNSC, LIHC, PAAD, and CHOL (P < .05) (Fig. 4C).

Figure 4.

The association between ATAD2 and clinical features in pan-cancer patient. (A) Correlation analysis examining the association between ATAD2 expression levels and patient gender. (B) Correlation analysis examining the association between ATAD2 expression levels and clinical stage. (C) Correlation analysis examining the association between ATAD2 expression levels and clinical grade. The symbols ns, *, **, ***, and **** represent statistical significance as follows: not significant, P < .05, P < .01, P < .001, and P < .0001 respectively. ATAD2 = ATPase Family AAA Domain Containing 2.

3.4. GSEA in pan-cancer

To elucidate the potential function and mechanism of ATAD2 in pan-cancer, we used GSEA to enrich ATAD2-related biological processes and KEGG pathways. The GSEA results for biological processes indicated that ATAD2 plays a role in several critical biological functions, including protein methylation, telomere maintenance, mRNA processing, and ribosome biogenesis (Fig. 5A, Fig. S7A–D, Supplemental Digital Content, https://links.lww.com/MD/P456). As depicted in Fig. 5B, our study revealed significant enhancements in several pathways critical to oncology, including those involved in the cell cycle, mismatch repair, spliceosome, pathways in cancer, DNA replication and P53 signaling pathway (Fig. 5B, Fig. S7E–H, Supplemental Digital Content, https://links.lww.com/MD/P456). Besides, to delve deeper into the role of ATAD2, we constructed a protein–protein interaction (PPI) network using data retrieved from the comPPI database. Within the confines of the nucleus, ATAD2 forms interactions with an ensemble of 22 unique proteins, predominantly including key players such as MYC, E2F2, E2F3, POLE3, and TRIM25, among others (Fig. 5C).

Figure 5.

GSEA and protein interaction analysis of ATAD2. (A) GSEA-based heatmap showing biological processes associated with ATAD2. (B) GSEA-based heatmap showing the Kyoto Encyclopedia of Genes and Genomes pathways associated with ATAD2. (C) ATAD2 protein network interaction diagram based on comPPI database. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2, GSEA = gene set enrichment analysis.

3.5. Gene alteration analysis in pan-cancer

To shed light on the mechanism contributing to the increased expression of ATAD2, we performed a CNV analysis on various samples. According to the analysis, most tumor types exhibit ATAD2 gene amplification in the majority of samples, with a small number of samples showing copy number deletions (Fig. 6A). Significant negative correlations were observed at the epigenetic level between the methylation levels of several sites in the ATAD2 promoter region and gene body and mRNA expression levels in most tumors, such as BLCA, BRCA, COAD, lymphoid neoplasm diffuse large B-cell lymphoma (DLBC), LIHC, SARC, SKCM, STAD, UCS and UVM (Fig. 6B). Furthermore, we explored the association between ATAD2 expression and MSI/TMB across different cancer types using data from TCGA. The findings demonstrated a significant positive relationship between ATAD2 expression and TMB in ACC, BLCA, BRCA, KICH, KIRP, LGG, LUAD, LUSC, PAAD, PRAD, SARC, STAD, and thymoma (THYM) (P < .05) as shown in Figure 6C. Additionally, our analysis revealed a positive association between ATAD2 expression and MSI in diverse cancer types, such as DLBC, GBM, KIRC, LUSC and STAD (P < .05) (Fig. 6D).

Figure 6.

Gene alteration analysis of ATAD2. (A) DNA copy number variation analysis of ATAD2 in pan-cancer. (B) DNA methylation analysis of ATAD2 in pan-cancer. (C) Radar plots depict the relationship between ATAD2 expression levels and TMB in pan-cancer. (D) Radar plots depict the relationship between ATAD2 expression levels and MSI in pan-cancer. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2, TMB = tumor mutation burden.

3.6. Immune infiltration analysis in pan-cancer

To elucidate its pan-cancer immunological significance, we analyzed the correlation between ATAD2 expression and immune cell infiltration. The results, based on the xCell algorithm, demonstrate a negative correlation between ATAD2 expression and the infiltration of multiple immune cell types in various cancer types, indicating a potential immunosuppressive effect of ATAD2 (Fig. 7A). The research indicated that CD4 + T helper 2 (Th2) cells exhibited a positive correlation with ATAD2 expression, suggesting a potential role for ATAD2 in modulating Th2-mediated immune responses (Fig. 7B). Conversely, both the stromal and microenvironment scores showed a negative correlation with ATAD2 expression across different cancer types, implying that higher ATAD2 levels may be associated with compromised tumor stroma and microenvironmental conditions (Fig. 7C and D).

Figure 7.

Immune infiltration analysis of ATAD2. (A) The heatmap presents the correlation between ATAD2 levels and diverse immune cell infiltration based on the xCell algorithm. Green to red gradient indicates correlation coefficient. Using the xCell algorithm, scatter plots illustrate the spearman correlation of ATAD2 expression with (B) CD4⁺ Th2 cells, (C) stroma score, and (D) microenvironment score in pan-cancer. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2.

3.7. Immune-related genes correlation analysis in pan-cancer

In an effort to investigate the correlation between ATAD2 expression and immune-related genes, we performed a correlation analysis between ATAD2 and a panel of immune-related genes, encompassing chemokines, chemokine receptors, immunostimulators, immunoinhibitors, and MHC genes (Fig. 8A). We found that ATAD2 expression is positively correlated with a wide range of immune-related genes, including TGFBR1, PVR, MICB, CD276, IL10RB, PDCD1, CD274, CCR8, and TAP2, suggesting a potential role in immune regulation. Notably, ATAD2 expression was found to be significantly correlated with immune-related genes in KIRC, OV, and UVM, highlighting the potential importance of ATAD2 in modulating immune responses in these cancer types. As shown in Figure 8B, a significant positive correlation is observed between ATAD2 and CD274 (PD-L1) expression in the majority of tumor types, including BLCA, KIRC, KICH, PRAD, OV, pheochromocytoma and paraganglioma (PCPG), DLBC, PAAD, CHOL, and UVM. Furthermore, we found an evident positive correlation between the expression levels of ATAD2 and PDCD1 in THYM, UVM, CHOL, KIRC, and LIHC tumors (Fig. 8C). According to correlation analysis, ATAD2 is positively associated with tumor RNA expression-based stemness score (RNAss) in the majority of tumors (Fig. 8D).

Figure 8.

Immune-related genes correlation analysis of ATAD2. (A) The heatmap illustrates the relationship between ATAD2 expression and immune-related genes. Blue to red gradient indicates correlation coefficient. Scatter plots illustrate the Spearman correlation of ATAD2 expression with (B) CD274 and (C) PDCD1 in pan-cancer. (D) Scatter plots show the results of Spearman correlation analysis between the expression of ATAD2 and tumor stemness score by using the RNAss algorithm. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2.

3.8. Therapeutic potential analysis of ATAD2

To evaluate the potential of ATAD2 as a predictive biomarker in immunotherapy, we analyzed 2 cohorts of patients with KIRC treated with anti-PD-1 and anti-PD-L1 therapies. As depicted in Figure 9A and B, elevated ATAD2 expression was associated with significantly worse survival outcomes in both groups. Similarly, in patients with SKCM receiving adoptive T-cell therapy, high ATAD2 expression corresponded to poor overall survival (Fig. 9C). To further investigate the role of ATAD2 in cancer therapy, we examined its impact on various treatment modalities, including chemotherapy and targeted therapies. Figure 9D–F demonstrates that elevated ATAD2 expression was associated with poorer survival outcomes in breast cancer and NSCLC patients. Additionally, in chemotherapy-resistant ovarian cancer, KM survival analysis revealed that high ATAD2 expression was linked to significantly worse overall survival (Fig. 9I). In liver cancer patients treated with transarterial chemoembolization (TACE), ATAD2 expression was significantly higher in non-responders, with the high-expression group showing a lower response rate than the low-expression group (Fig. 9G). Distinctly, in liver cancer patients treated with sorafenib, high ATAD2 expression was observed in non-responders, who also exhibited a notably lower response rate compared to those with low ATAD2 expression (Fig. 9H).

Figure 9.

Therapeutic analysis of ATAD2. (A) Kaplan–Meier survival analysis for KIRC patients in the PMID32472114 cohort treated with anti-PD-1 therapy. (B) Kaplan–Meier survival analysis for KIRC patients in the PMID32895571 cohort treated with anti-PD-L1 therapy. (C) Kaplan–Meier survival analysis for SKCM patients undergoing adoptive T-cell therapy (ACT) in the GSE100797 cohort. (D and E) Kaplan–Meier survival analysis for BRCA patients treated with chemotherapy in the GSE25055 and GSE25065 cohorts. (F) Kaplan–Meier survival analysis for NSCLC patients treated with chemotherapy in the GSE42127 cohort. (G) Boxplot (top) and proportion chart (bottom) demonstrating ATAD2 expression differences and response distributions in LIHC patients undergoing transarterial chemoembolization (TACE), based on the GSE104580 dataset. (H) Boxplot (top) and proportion chart (bottom) demonstrating ATAD2 expression differences and response distributions in LIHC patients undergoing sorafenib treatment, based on the GSE109211 dataset. (I) Kaplan–Meier survival analysis for OV patients with chemotherapy resistance in the GSE63885 cohort. The symbols ns, *, **, and *** represent statistical significance as follows: not significant, P < .05, P < .01, and P < .001, respectively. ATAD2 = ATPase Family AAA Domain Containing 2, BRCA = breast invasive carcinoma, DRFS = distant relapse-free survival, KIRC = kidney renal clear cell carcinoma, LIHC = liver hepatocellular carcinoma, NSCLC = non-small cell lung cancer, OS = overall survival, OV = ovarian serous cystadenocarcinoma, SKCM = skin cutaneous melanoma.

3.9. Immunohistochemical staining

To validate the differential expression of ATAD2 in normal and tumor tissues, immunohistochemistry-stained images were obtained from HPA database. As demonstrated in Figure 10, ATAD2 protein levels are markedly higher in 12 types of cancer compared to normal tissues.

Figure 10.

Immunohistochemical staining of ATAD2. Representative images of immunohistochemical staining of ATAD2 in 12 types of normal and tumor tissues. ATAD2 = ATPase Family AAA Domain Containing 2.

4. Discussion

The combined dysregulation of genetic and epigenetic factors represents a universal hallmark of cancer, driving uncontrolled cellular division and proliferation even in adverse environments.^[30] Epigenetic regulatory factors exert profound influence over tumorigenesis, progression, metastasis, immune evasion, treatment response, and drug resistance, establishing them as promising therapeutic targets.^[31] ATAD2 is a key chromatin-modifying factor that has been implicated in the pathogenesis of cancer.^[32] Studies have reported that oncogenic competence is a critical determinant of cancer formation. Furthermore, ATAD2 has been found to play a crucial role in enhancing oncogenic competence and is indispensable for the formation of zebrafish melanoma.^[33]

This study provides an in-depth exploration of the multidimensional role of the ATAD2 gene in cancer by integrating data from multiple databases. By examining normal tissues in the GTEx database, we discovered that the expression pattern of ATAD2 is strongly linked to its function in DNA replication. It is highly expressed in proliferative tissues like bone marrow and testis, whereas its expression is lower in tissues with slower cell division, such as the kidneys and blood. This expression pattern might be connected to its role in cell cycle and DNA replication regulation, which is consistent with the known function of ATAD2 as a chromatin remodeling regulatory factor.^[34] Across TCGA samples, TCGA GTEx samples, and TCGA paired samples, we found significant differences in ATAD2 expression when comparing tumor and normal tissues. With the exception of THYM and TGCT, ATAD2 mRNA expression was markedly higher in most cancers relative to their normal tissue counterparts. Notably, our protein-level analysis demonstrated that ATAD2 protein is upregulated in a diverse range of 12 cancers, encompassing BRCA, KIRC, COAD, GBM, LIHC, HNSC, LUSC, LUAD, OV, PAAD, and UCEC. This finding was further validated by immunohistochemistry images available in the HPA database. The findings presented above indicate that ATAD2 could be a key player in the development of these malignancies, possibly functioning as an oncogene. Notwithstanding the significant advancements in our research, certain limitations persist. For instance, ATAD2 mRNA levels exhibit significant downregulation in TYHM and TGCT. This phenomenon may be attributed to the diversity in sample origins and the variations in technological platforms (such as sequencing or microarrays) across different databases. Consequently, to improve the precision of our results, we propose expanding the control group sample size, thereby reducing the probability of such discrepancies.

To date, no extensive research has been conducted to evaluate the prognostic value of ATAD2 in a wide range of cancers. Our study sought to bridge this gap by exploring the multifaceted prognostic implications of ATAD2 overexpression on OS in various malignancies through comprehensive analyses of TCGA and GEO databases. Our findings suggest that increased ATAD2 expression is linked to poorer outcomes in terms of OS, DSS, PFI, and DFI, particularly in cancers such as ACC, KICH, KIRP, LGG, LUAD, MESO, PAAD, and UVM. These findings are further corroborated by existing literature.^{[10,14,35,36]} ROC curves have become a crucial tool in cancer diagnostics, providing accurate assessments of test performance and guiding cancer detection and treatment decisions.^[37,38] The present study reveals a strong association between ATAD2 overexpression and diagnostic accuracy in 17 cancer types, highlighting its potential value as a diagnostic marker. For clinical significance, we observed an association between ATAD2 expression and clinical stage, grade, and gender in some cancers. This suggests a potential role for ATAD2 in disease progression and the regulation of sex hormones.

Functionally, GSEA results indicate that ATAD2 was implicated in numerous cancer-related biological processes and pathways. Specifically, ATAD2 played a crucial role in regulating cell cycle and DNA replication. Mechanistically, the PPI network revealed that ATAD2 interacts with transcription factor E2F1, oncogene MYC, and ubiquitin ligase TRIM25 within the nucleus. The finding was validated in a study conducted by Tong et al.^[12] Methylation typically functions to modulate gene expression under normal conditions. However, in tumors, abnormal methylation events can lead to the silencing of tumor suppressor genes, activation of oncogenes, and facilitate tumor progression.^[39,40] The correlation analysis revealed a significant negative correlation between the expression level of ATAD2 mRNA and the methylation levels of multiple sites in vivo. CNVs are strongly implicated in cancer genome instability and represent a fundamental driver of cancer development.^[41] It was found that in the bulk of tumor samples, there was a notable presence of copy number amplification. Consequently, we posit that the elevated expression of ATAD2 mRNA is likely modulated by copy number amplification and altered methylation levers.

Evidence suggests that patients exhibiting a higher TMB tend to respond better to immunotherapy.^[42] Combining TMB and MSI assessments allows for a more precise prediction of tumor response to immunotherapy.^[43] It has been discovered that the expression of ATAD2 showed a significant correlation with TMB and MSI in most tumors, including but not limited to LUSC, LUAD, BLCA, and KIRC. Uncontrolled immune cell proliferation and dysfunction are key contributors to tumor development.^[44] Our findings indicate that ATAD2 expression was linked to the infiltration of multiple immune cell types in most tumors, including CD4 + T cells, CD8 + T cells, B cells, macrophages, and cancer-associated fibroblasts. Cancer-associated fibroblasts are deeply involved in tumor occurrence, development, metastasis, and immune therapy response.^[45,46] Cancer stem cells, with their distinctive capacity for self-renewal and differentiation, significantly contribute to the high rates of tumor recurrence and metastasis, profoundly influencing the effectiveness of treatment.^[47,48] To a certain extent, RNAss can indicate the stem cell properties of tumors. In our research, we found that ATAD2 expression was positively associated with the RNAss in the majority of tumors studied. Besides, ATAD2 was significantly correlated with the expression of multiple immune-related genes, particularly immune checkpoint genes such as PDCD1, CD274, and CTLA4. In our study, it was found that ATAD2 expression was positively correlated with PD-L1 expression in BLCA and KIRC. Immune checkpoint blockade therapy, particularly anti-PD1/PD-L1 therapy, has been proven effective in improving the prognosis of patients with advanced or metastatic renal cell carcinoma.^[49] Our study revealed that, in patients with advanced RCC undergoing anti-PD-1/PD-L1 treatment, a higher level of ATAD2 expression is correlated with decreased survival outcomes. Notably, in SKCM patients treated with adoptive T-cell therapy (ACT), a strategy known to enhance anti-tumor immune responses, elevated ATAD2 expression was associated with poor overall survival.^[50] Targeting ATAD2 may boost the effectiveness of chemotherapy by overcoming drug resistance.^[51] This is consistent with findings showing that, in chemotherapy-treated patients with BRCA, NSCLC, and OV, elevated ATAD2 expression is linked to reduced therapeutic responses and unfavorable survival outcomes. In LIHC patients receiving TACE or sorafenib, elevated ATAD2 expression was associated with reduced response rates and adverse survival outcomes. By linking ATAD2 to therapeutic resistance, these findings propose its dual role as a biomarker for predicting poor treatment outcomes and as a potential therapeutic target. These results are supported by previous studies suggesting that ATAD2 contributes to therapeutic resistance by promoting cell cycle progression and enhancing DNA repair capabilities.^[52–54] Additionally, its role in shaping the tumor immune microenvironment warrants further investigation, as it may underlie its association with poor immunotherapy responses. Moreover, targeting ATAD2 with small-molecule inhibitors represents a promising strategy to enhance therapeutic efficacy, especially in resistant cancers.^[55]

These findings highlight ATAD2’s significant potential as both a predictive biomarker and a therapeutic target across multiple cancer types. Assessing ATAD2 expression levels prior to treatment could aid in patient stratification, enabling the selection of optimal therapeutic regimens and identifying individuals at higher risk for drug resistance. Moreover, the development of ATAD2-targeted inhibitors – either as monotherapy or in combination with chemotherapy, immunotherapy, or other targeted therapies – offers promising strategies to overcome resistance and improve clinical outcomes. However, this study is constrained by its reliance on retrospective datasets, underscoring the need for validation through prospective clinical trials. Additionally, the precise molecular mechanisms underlying ATAD2-driven therapeutic resistance remain incompletely understood and warrant further investigation. Future research should prioritize the development of ATAD2-specific therapies and explore its potential in combination treatments, particularly with immune checkpoint inhibitors, to enhance therapeutic efficacy and address resistance challenges.

5. Conclusion

Our study comprehensively analyzed ATAD2 expression across multiple cancer types. We identified that ATAD2 is significantly upregulated in various malignant tumors and consistently associated with poor prognosis across multiple datasets. Additionally, ATAD2’s involvement in chemotherapy, immunotherapy, and targeted therapy response underscores its therapeutic significance. The observed correlations between ATAD2 expression, immune pathway activity, tumor immune infiltration, and resistance-related mechanisms further highlight its role in cancer progression. While further experimental validation and clinical trials are needed, our findings provide valuable insights into the role of ATAD2 in cancer and suggest its potential as both a diagnostic and therapeutic target.

Acknowledgments

The authors acknowledge the invaluable support from public databases, websites, and software used in this paper.

Author contributions

Conceptualization: Shijie Liang, Haodong Liu.

Data curation: Haodong Liu.

Formal analysis: Qisheng Su.

Funding acquisition: Zheng Yang.

Investigation: Qisheng Su.

Methodology: Shijie Liang.

Supervision: Shijie Liang.

Visualization: Qisheng Su.

Writing – original draft: Zheng Yang.

Writing – review & editing: Wuning Mo.