ELSEVIER
Cancer Genetics
journal homepage: www.elsevier.com/locate/cancergen
Cancer Genetics
-
-
C
A
Check for updates
Synergistic dual oncogenic role of CKS2 and ACAT2: Enhanced regulatory coherence and biomarker potential in adrenocortical carcinoma
Javad Omidi İD
Department of Chemical Engineering, Columbia University, NY, NY 10027, USA
ARTICLE INFO
Keywords:
Adrenocortical carcinoma Cerna network Oncogenic biomarkers Transcriptional co-regulation
ABSTRACT
Adrenocortical carcinoma (ACC) is a rare and highly aggressive endocrine malignancy with limited therapeutic options and poor clinical outcomes. To identify molecular biomarkers with diagnostic and prognostic relevance, an integrative ceRNA network analysis was performed using transcriptomic profiles from TCGA-ACC and GTEx adrenal tissues. miR-466, the highest-centrality miRNA in the ceRNA network, was used as the focal regulatory component in this framework. Experimentally validated targets were obtained from miRTarBase and miRDB, refined by TargetScan binding-site specificity and correlation-based filtering, and evaluated for tumor-specific expression. Among these candidates, CKS2 and ACAT2 emerged as consistently and significantly upregulated across therapy subgroups, with expression patterns confirmed across patient-level clinical heterogeneity. Both genes demonstrated markedly elevated expression in ACC relative to normal adrenal tissue and exhibited rein- forced transcriptional co-regulation in tumors, suggesting a coordinated regulatory shift. Kaplan-Meier survival analyses indicated that high expression of either gene was associated with reduced overall survival, while a multivariate CoxPH model integrating both markers stratified patients into distinct high- and low-risk groups. Additionally, ROC classification demonstrated strong diagnostic performance for distinguishing tumor from normal tissue (AUC = 0.90 for CKS2, 0.88 for ACAT2, and 0.90 for the combined logistic regression model). Functional annotation revealed that CKS2 regulates cyclin-dependent cell cycle transitions, while ACAT2 par- ticipates in lipid and fatty-acid metabolic pathways. Together, these roles support a synergistic oncogenic axis in which proliferative acceleration is metabolically sustained, reinforcing tumor growth. These findings nominate CKS2 and ACAT2 as robust biomarkers and mechanistic drivers with translational relevance in ACC.
Background
Adrenocortical carcinoma (ACC) is a rare but highly aggressive malignancy of the adrenal cortex, associated with poor clinical out- comes despite advances in surgical and systemic therapies [1-4]. The five-year survival rate for advanced disease remains below 40 %, largely due to late diagnosis, high recurrence rates, and limited responsiveness to current treatment strategies [3,4]. These challenges underscore the need for reliable molecular biomarkers capable of improving early detection, prognostic assessment, and therapeutic decision-making in ACC [5,6].
Recent progress in cancer genomics has highlighted the crucial reg- ulatory roles of non-coding RNAs (ncRNAs), particularly microRNAs (miRNAs) and long non-coding RNAs (lncRNAs), in modulating onco- genic pathways across diverse tumor types, including ACC [5,7,8]. The competing endogenous RNA (ceRNA) hypothesis provides a framework through which transcripts harboring shared miRNA response elements
exert reciprocal regulatory effects, forming interconnected layers of post-transcriptional control [9,10]. Dysregulation of ceRNA interactions has been shown to promote tumor cell proliferation, metastasis, and therapeutic resistance, suggesting that ceRNA network perturbations may reveal key molecular drivers and clinically informative biomarkers.
Systems biology and network-based analytical approaches have enabled the integration of multi-omics data to identify central regula- tory nodes within complex gene expression networks [11-14]. While such strategies have successfully uncovered diagnostic and prognostic biomarkers in several cancers, comprehensive ceRNA network analyses in ACC remain limited, and the functional relevance of individual reg- ulatory components has yet to be fully elucidated [5]. The availability of large-scale transcriptomic resources, including TCGA-ACC and recent GTEx datasets [15,16], now provides an opportunity to systematically characterize ceRNA interactions and identify novel molecular signatures with potential diagnostic and therapeutic value in ACC.
E-mail address: jo2668@columbia.edu.
https://doi.org/10.1016/j.cancergen.2025.10.103
Objectives
This study was designed to identify clinically relevant molecular biomarkers in ACC through an integrative ceRNA network-based framework. Transcriptomic profiles from TCGA-ACC [15] and GTEx-2025 [16] normal adrenal tissues (Table S1) were preprocessed with batch correction, normalization, and expression filtering to ensure cross-cohort comparability. Experimentally validated miRNA-mRNA interactions were incorporated, and network topology metrics were applied to prioritize central regulatory nodes. Differential expression, co-expression structure, and survival association analyses were
conducted to refine candidate selection. The objective was to elucidate previously uncharacterized post-transcriptional regulation and nomi- nate molecular targets with diagnostic, prognostic, and therapeutic relevance for precision oncology in ACC.
Results
A systems-level ceRNA network was constructed from ACC tran- scriptomes, and network topology analysis identified hsa-miR-466 as the highest-centrality regulatory hub (Fig. 1a). Candidate target genes of miR-466 were retrieved from miRTarBase [17] and miRDB [18], and
300
250
-log10(p-value)
200
150
hsa-mir-466
100
50
(a)
(b)
0
-10
-5
0
5
10
log2(FC)
Low Centrality
· CHEK1
GOLGA7B
ACAT2
C4orf46
miRNA
BRIX1
LSM5
IDO1
CKS2
Medium Centrality
MELK
LYAR
ERP44
CDYL2
mRNA
PLK4
SUB1
High Centrality
MRPL42
ZNF562
O CENPI
HSBP1
DEPDC1
RAB3B
☒
NHLRC1
STC2
HOMER1
GJC1
Relative Expression Level per Therapy Group
Mean Expression (TPM)
600
(c)
Pharma+Radiation
Pharma only
400
All Cases
200
0
ERP44
DEPDC1
CENPI
LSM5
SUB1
BRIX1
STC2
ACAT2
CKS2
IDO1
PLK4
LYAR
CHEK1
HOMER1
GOLGA7B
MELK
CDYL2
RAB3B
ZNF562
GJC1
NHLRC1
MRPL42
C4orf46
HSBP1
intersection results are summarized in Table S2. These targets were next evaluated in ACC transcriptomic data, and only pairs displaying signif- icant negative correlation (p < 0.05) were retained. Upregulated genes were prioritized as putative oncogenic candidates. To refine specificity, TargetScan [19] filtering was applied, selecting transcripts containing 8mer or 7mer-m8 binding sites and Total context++ scores < - 0.1, yielding the final gene set shown in the volcano plot (Fig. 1b).
To assess clinical relevance, expression patterns of the selected target genes were compared across therapy subgroups within the TCGA-ACC cohort. Mean normalized expression values demonstrated consistent upregulation of CKS2 and ACAT2 across patients receiving pharma- ceutical therapy alone or in combination with radiation, indicating treatment-independent oncogenic activity (Fig. 1c). Other candidates
exhibited comparatively modest variability. A corresponding annotated heatmap illustrating patient-level and clinical subgroup heterogeneity is provided in Figure S1, demonstrating that elevated expression of CKS2 and ACAT2 is consistently maintained across age, stage, and treatment categories.
Fig. 2a demonstrates that CKS2 and ACAT2 are markedly upregu- lated in ACC tumor tissues relative to normal adrenal samples (p < 0.001), consistent with their proposed oncogenic roles. Moreover, a pronounced increase in CKS2-ACAT2 transcriptional correlation is observed specifically in tumor samples (Fig. 2b), indicating reinforced regulatory coherence that is absent in normal tissue. This coordinated upregulation and strengthened coupling support a synergistic dual oncogenic axis, highlighting their potential as clinically relevant
CKS2
(a)
10
12
ACAT2
3.97e-23
4.98e-07
(b)
CKS2 vs. ACAT2
9
11
Normal (r=0.03)
Log2 (Expression)
8
Tumor (r=0.44)
8
10
log2(ACAT2 TPM + 0.01)
9
7
8
6
6
7
4
5
6
5
2
4
4
3
0
3
2
2
4
6
8
1
Normal
Tumor
1
Normal
Tumor
log2(CKS2 TPM + 0.01)
(c)
CKS2
ACAT2
Multi-gene Risk Model
1.0
1.0
Low Group
1.0
High Group
Percent Survival
0.8
0.8
0.8
0.6
0.6
0.6
0.4
Logrank p = 1.1e-06
0.4
HR(high) =4.49
Logrank p = 6.4e-02
0.4
HR(high) = 2.06
Logrank p = L.1e-06
n(high) = 40, n(low) = 39
n(high) = 40, n(low) = 39
HR(high vs low) =4.49
0.2
Median High = 39.9m [0.1-1.0]
0.2
Median High = 70.2m [0.2-1.0]
0.2
n(high) = 40, n(low) - 39
Median High = 39.9m [0.1-1.0]
0.0
Median Low = infm [0.5-1.0]
0.0
Median Low = infm [0.4-1.0]
Median Low = infm [0.5-1.0]
0
50
100
150
0
50
100
150
0.0
0
50
100
150
Months
Months
Months
(d)
CKS2
ACAT2
Multi-Gene LR Model
1.0
1.0
1.0
True Positive Rate
0.8
0.8
0.8
0.6
0.6
0.6
0.4
0.4
0.4
AUC = 0.90
AUC = 0.88
AUC = 0.90
0.2
p = 2.65e-27
0.2
p = 8.79e-54
0.2
p = 2.61e-27
0.0
0.0 0.2 0.4 0.6 0.8 1.0
0.0
).0 0.2 0.4 0.6 0.8 1.0
0.0
0.0
0.2
0.4
0.6
0.8
1.0
False Positive Rate
False Positive Rate
False Positive Rate
molecular drivers in ACC.
Kaplan-Meier analysis demonstrated that high expression of CKS2 and ACAT2 was associated with significantly reduced overall survival in ACC patients (Fig. 2c). A multivariate CoxPH model integrating these oncogenic candidates further stratified patients into high- and low-risk groups based on median risk score, revealing a similarly adverse sur- vival pattern. In parallel, ROC analysis showed strong discriminatory power for distinguishing tumor from normal tissue, with AUCs of 0.90 (CKS2), 0.88 (ACAT2), and 0.90 for the multi-gene logistic regression model (Fig. 2d), confirming both prognostic and diagnostic utility.
Functional annotation revealed that CKS2 is linked to regulatory complexes controlling cell cycle progression, including association with cyclin-dependent kinase holoenzymes, SCF ubiquitin ligase components, and factors involved in chromatin interaction and mitotic phase tran- sitions (Table S3). In contrast, ACAT2 was annotated to pathways gov- erning lipid and fatty-acid metabolism, including acetyl-CoA conversion, fatty-acid degradation, and broader central metabolic signaling networks essential for energy flux (Table S4). These annota- tions indicate that CKS2 promotes proliferative drive while ACAT2 supports metabolic reprogramming, two fundamental hallmarks of tumor progression. Their concurrent upregulation therefore suggests a synergistic oncogenic axis, whereby accelerated cell cycle activity is metabolically sustained through lipid-derived energy production, collectively reinforcing tumor growth and survival in ACC [20,21].
Conclusion
This study identified CKS2 and ACAT2 as central molecular drivers in ACC through an integrative ceRNA network framework, demonstrating their consistent upregulation, reinforced transcriptional coupling, adverse survival associations, and strong diagnostic performance. Functional annotations further revealed that CKS2 promotes prolifera- tive cell-cycle signaling, while ACAT2 supports lipid-dependent meta- bolic reprogramming, together forming a synergistic oncogenic axis that may sustain the energetic and proliferative demands of ACC progression. These findings highlight both genes as promising biomarkers with po- tential utility in risk stratification and therapeutic targeting.
However, several limitations should be acknowledged. The analyses were constrained by the limited sample size of TCGA-ACC, which may affect statistical robustness and generalizability. Additionally, the lack of in vitro or in vivo functional validation precludes definitive causal inference regarding their mechanistic roles. Future work incorporating larger, clinically diverse cohorts and experimental validation will be essential to fully establish the translational and therapeutic relevance of this oncogenic axis in ACC.
Funding statement
No specific grant was received for this research.
Ethics statement
N/A.
Conflict of interest statement
The author declares no conflicts of interest.
Declaration of generative AI and AI-assisted technologies in the writing process
During the preparation of this work the author used ChatGPT (OpenAI) to improve English language and readability. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.
CRediT authorship contribution statement
Javad Omidi: Writing - review & editing, Writing - original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Supplementary materials
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cancergen.2025.10.103.
Data availability
Data is available from the corresponding author upon reasonable request.
References
[1] Allolio Bruno, Fassnacht Martin. Adrenocortical carcinoma: clinical update. J Clin Endocrinol Metab 2006;91(6):2027-37.
[2] Kerkhofs Thomas MA, Verhoeven Rob HA, Zwan Jan Maarten Van der, Dieleman Jeanne, Kerstens Michiel N, Links Thera P, Poll-Franse Lonneke VVan de, Haak Harm R. Adrenocortical carcinoma: a population-based study on incidence and survival in the Netherlands since 1993. Eur J Cancer 2013;49(11):2579-86.
[3] Terzolo Massimo, Angeli Alberto, Fassnacht Martin, Daffara Fulvia, Tauchmanova Libuse, Conton Pier Antonio, Rossetto Ruth, et al. Adjuvant mitotane treatment for adrenocortical carcinoma. N Engl J Med 2007;356(23):2372-80.
[4] Fassnacht Martin, Libé Rossella, Kroiss Matthias, Allolio Bruno. Adrenocortical carcinoma: a clinician’s update. Nat Rev Endocrinol 2011;7(6):323-35.
[5] Decmann Abel, Perge Pál, Turai Peter Istvan, Patócs Attila, Igaz Peter. Non-coding RNAs in adrenocortical cancer: from pathogenesis to diagnosis. Cancers (Basel) 2020;12(2):461.
[6] McBrearty Noreen, Bahal Devika, Platero Suso. Fast-tracking drug development with biomarkers and companion diagnostics. J Cancer Metastasis Treat 2024;10. N- A.
[7] Slack Frank J, Chinnaiyan Arul M. The role of non-coding RNAs in oncology. Cell 2019;179(5):1033-55.
[8] Salimian Niloufar, Peymani Maryam, Ghaedi Kamran, Mirzaei Sepideh, Hashemi Mehrdad. Diagnostic and therapeutic potential of LINC01929 as an oncogenic LncRNA in human cancers. Pathol-Res Pract 2023;244:154409.
[9] Tay Yvonne, Rinn John, Pandolfi Pier Paolo. The multilayered complexity of ceRNA crosstalk and competition. Nature 2014;505(7483):344-52.
[10] Yang Ni, Liu Kuo, Yang Mengxuan, Gao Xiang. ceRNAs in cancer: mechanism and functions in a comprehensive regulatory network. J Oncol 2021;2021(1):4279039.
[11] Vidal Marc, Cusick Michael E, Barabási Albert-László. Interactome networks and human disease. Cell 2011;144(6):986-98.
[12] Qi Xin, Lin Yuxin, Chen Jiajia, Shen Bairong. Decoding competing endogenous RNA networks for cancer biomarker discovery. Br Bioinform 2020;21(2):441-57.
[13] Li Xing, Li Bing, Ran Pixin, Wang Lanying. Identification of ceRNA network based on a RNAseq shows prognostic lncRNA biomarkers in human lung adenocarcinoma. Oncol Lett 2018;16(5):5697-708.
[14] Omidi Javad. miR-507 and miR-665 as central MicroRNA regulators in the ceRNA network of adrenocortical carcinoma: a systems biology approach. Hum Gene 2025;46:201498.
[15] The Cancer Genome Atlas (TCGA). TCGA Program. Natl Cancer Inst 2024. http s://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tc ga.
[16] GTEx 2025 Consortium Ardlie Kristin G, Deluca David S, Segrè Ayellet V, Sullivan Timothy J, Young Taylor R, Gelfand Ellen T, et al. The Genotype-Tissue Expression (GTEx 2025) pilot analysis: multitissue gene regulation in humans. Science 2015;348(6235):648-60.
[17] Huang Hsi-Yuan, Lin Yang-Chi-Dung, Li Jing, Huang Kai-Yao, Shrestha Sirjana, Hong Hsiao-Chin, Tang Yun, et al. miRTarBase 2020: updates to the experimentally validated microRNA-target interaction database. Nucleic Acids Res 2020;48(D1): D148-54.
[18] Chen Yuhao, Wang Xiaowei. miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Res 2020;48(D1):D127-31.
[19] Agarwal Vikram, Bell George W, Nam Jin-Wu, Bartel David P. Predicting effective microRNA target sites in mammalian mRNAs. elife 2015;4:e05005.
[20] UniProt: the universal protein knowledgebase in 2025. Nucleic Acids Res 2025;53 (D1):D609-17.
[21] Kanehisa Minoru, Furumichi Miho, Sato Yoko, Matsuura Yuriko, Ishiguro- Watanabe Mari. KEGG: biological systems database as a model of the real world. Nucleic Acids Res 2025;53(D1):D672-7.