Identification of Prognostic Biomarkers in Gene Expression Profile of Neuroblastoma Via Machine Learning

Abstract

Neuroblastoma (NB) is a common pediatric solid malignancy characterized by heterogeneous clinical outcomes. The identification of predictive and interpretable prognostic biomarkers is critical for advancing precision medicine in NB. We proposed an integrative network-based machine learning method for biomarker discovery, which employed a network smoothed t-statistic support vector machine to select prognostic related biomarkers, and then we performed network analysis on these biomarkers to find hub genes. Later, we conducted a comprehensive analysis to integrate bulk and single-cell RNA sequencing data to character the tumor microenvironment of prognostic state and correlated them to the discovered hub genes. This analysis identified 528 prognostic biomarkers associated with NB. Network-based analysis further refined this set to 11 hub prognostic biomarkers for NB: AURKA, BLM, BRCA1, BRCA2, CCNA2, CHEK1, E2F1, MAD2L1, PLK1, RAD51, and RFC3. Among these genes, high RFC3 expression was significantly associated with poor prognosis, highlighting its potential as a novel prognostic biomarker in NB. Additionally, our findings revealed that these biomarkers are correlated to chemotherapy drugs, such as vincristine and cyclophosphamide. Furthermore, drug sensitivity analyses identified several candidate drugs, such as dactinomycin, bortezomib, docetaxel, and sepantronium bromide, that may hold therapeutic potential for NB treatment. This study offers novel insights to underlying NB prognosis and therapeutic targets and provides a foundation for developing personalized treatment strategies to improve clinical outcomes.

Keywords

gene expression profile / machine learning / molecular network / neuroblastoma / prognostic biomarkers

References

[1] J. M. Maris, “Recent Advances in Neuroblastoma,” New England Journal of Medicine 362, no. 23 (2010): 2202–2211.

[2] I. Janoueix-Lerosey, G. Schleiermacher, and O. Delattre, “Molecular Pathogenesis of Peripheral Neuroblastic Tumors,” Oncogene 29, no. 11 (2010): 1566–1579.

[3] C. C. Swift, M. J. Eklund, J. M. Kraveka, and A. L. Alazraki, “Updates in Diagnosis, Management, and Treatment of Neuroblastoma,” RadioGraphics 38, no. 2 (2018): 566–580.

[4] J. Wienke, L. L. Visser, W. M. Kholosy, et al., “Integrative Analysis of Neuroblastoma by Single-Cell RNA Sequencing Identifies the NECTIN2-TIGIT Axis as a Target for Immunotherapy,” Cancer Cell 42, no. 2 (2024): 283–300.e8.

[5] N. L. Henry and D. F. Hayes, “Cancer Biomarkers,” Molecular Oncology 6, no. 2 (2012): 140–146.

[6] S. Ng, S. Masarone, D. Watson, and M. R. Barnes, “The Benefits and Pitfalls of Machine Learning for Biomarker Discovery,” Cell and Tissue Research 394, no. 1 (2023): 17–31.

[7] M. Huang and W. A. Weiss, “Neuroblastoma and Mycn,” Cold Spring Harbor Perspectives in Medicine 3, no. 10 (2013): a014415.

[8] D. Trochet, F. Bourdeaut, I. Janoueix-Lerosey, et al., “Germline Mutations of the Paired-Like Homeobox 2B (PHOX2B) Gene in Neuroblastoma,” American Journal of Human Genetics 74, no. 4 (2004): 761–764.

[9] E. L. Carpenter and Y. P. Mossé, “Targeting ALK in Neuroblastoma: Preclinical and Clinical Advancements,” Nature Reviews Clinical Oncology 9, no. 7 (2012): 391–399.

[10] M. Peifer, F. Hertwig, F. Roels, et al., “Telomerase Activation by Genomic Rearrangements in High-Risk Neuroblastoma,” Nature 526, no. 7575 (2015): 700–704.

[11] C. Wang, B. Gong, P. R. Bushel, et al., “The Concordance Between RNA-Seq and Microarray Data Depends on Chemical Treatment and Transcript Abundance,” Nature Biotechnology 32, no. 9 (2014): 926–932.

[12] A. Oberthuer, D. Juraeva, B. Hero, et al., “Revised Risk Estimation and Treatment Stratification of Low- and Intermediate-Risk Neuroblastoma Patients by Integrating Clinical and Molecular Prognostic Markers,” Clinical Cancer Research 21, no. 8 (2015): 1904–1915.

[13] X. Yuan, J. A. Seneviratne, S. Du, et al., “Single-Cell Profiling of Peripheral Neuroblastic Tumors Identifies an Aggressive Transitional State That Bridges an Adrenergic-Mesenchymal Trajectory,” Cell Reports 41, no. 1 (2022): 111455.

[14] Y. Cun and H. Fröhlich, “Network and Data Integration for Biomarker Signature Discovery Via Network Smoothed T-Statistics,” PLoS One 8, no. 9 (2013): e73074.

[15] Y. Cun and H. Fröhlich, “NetClass: An R-Package for Network Based, Integrative Biomarker Signature Discovery,” Bioinformatics (Oxford, England) 30, no. 9 (2014): 1325–1326.

[16] M. E. Ritchie, B. Phipson, D. Wu, et al., “Limma Powers Differential Expression Analyses for RNA-Sequencing and Microarray Studies,” Nucleic Acids Research 43, no. 7 (2015): e47.

[17] Z. Gu, R. Eils, and M. Schlesner, “Complex Heatmaps Reveal Patterns and Correlations in Multidimensional Genomic Data,” Bioinformatics (Oxford, England) 32, no. 18 (2016): 2847–2849.

[18] P. Langfelder and S. Horvath, “WGCNA: An R Package for Weighted Correlation Network Analysis,” BMC Bioinformatics 9, no. 1 (2008): 559.

[19] D. Zhao, Y. Zhang, H. Ren, et al., “Multi-Omics Analysis Reveals the Evolutionary Origin of Diterpenoid Alkaloid Biosynthesis Pathways in Aconitum,” Journal of Integrative Plant Biology 65, no. 10 (2023): 2320–2335.

[20] F. Yang, J. Fan, R. Yang, and Y. Cun, “Integrative Analysis of Blood Transcriptome Profiles in Small-Cell Lung Cancer Patients for Identification of Novel Chemotherapy Resistance-Related Biomarkers,” Frontiers in Immunology 15 (2024): 1338162.

[21] P. Shannon, A. Markiel, O. Ozier, et al., “Cytoscape: A Software Environment for Integrated Models of Biomolecular Interaction Networks,” Genome Research 13, no. 11 (2003): 2498–2504.

[22] G. Yu, L.-G. Wang, Y. Han, and Q.-Y. He, “ClusterProfiler: An R Package for Comparing Biological Themes Among Gene Clusters,” OMICS 16, no. 5 (2012): 284–287.

[23] H. Wickham, “Stringr: Simple, Consistent Wrappers for Common String Operations,” R package version 1.5.1 (2023), https://github.com/tidyverse/stringr.

[24] V. Gómez-Rubio, “ggplot2 - Elegant Graphics for Data Analysis, 2nd edn,” Journal of Statistical Software 77, no. Book Review 2 (2017): 1–3.

[25] W. Walter, F. Sánchez-Cabo, and M. Ricote, “GOplot: An R Package for Visually Combining Expression Data With Functional Analysis,” Bioinformatics (Oxford, England) 31, no. 17 (2015): 2912–2914.

[26] G. Yu, “Enrichplot: Visualization of Functional Enrichment Result,” R package version 1.22.0 (2021), https://yulab-smu.top/biomedical-knowledge-mining-book/.

[27] R. Ietswaart, B. M. Gyori, J. A. Bachman, P. K. Sorger, and L. S. Churchman, “GeneWalk Identifies Relevant Gene Functions for a Biological Context Using Network Representation Learning,” Genome Biology 22, no. 1 (2021): 55.

[28] H. Lin and D. Zelterman, “Modeling Survival Data: Extending the Cox Model,” Technometrics 44, no. 1 (2002): 85–86.

[29] D. Aran, Z. Hu, and A. J. Butte, “XCell: Digitally Portraying the Tissue Cellular Heterogeneity Landscape,” Genome Biology 18, no. 1 (2017): 220.

[30] A. Kassambara, “Ggpubr: ‘ggplot2’ Based Publication Ready Plots,” R package version 0.6.0 (2023), https://rpkgs.datanovia.com/ggpubr/.

[31] K. Yoshihara, M. Shahmoradgoli, E. Martínez, et al., “Inferring Tumour Purity and Stromal and Immune Cell Admixture From Expression Data,” Nature Communications 4, no. 1 (2013): 2612.

[32] E. Becht, N. A. Giraldo, L. Lacroix, et al., “Estimating the Population Abundance of Tissue-Infiltrating Immune and Stromal Cell Populations Using Gene Expression,” Genome Biology 17, no. 1 (2016): 218.

[33] D. Maeser, R. F. Gruener, and R. S. Huang, “OncoPredict: An R Package for Predicting In Vivo or Cancer Patient Drug Response and Biomarkers From Cell Line Screening Data,” Briefings in Bioinformatics 22, no. 6 (2021): bbab260.

[34] Y. Hao, T. Stuart, M. H. Kowalski, et al., “Dictionary Learning for Integrative, Multimodal and Scalable Single-Cell Analysis,” Nature Biotechnology 42, no. 2 (2023): 293–304.

[35] C. S. McGinnis, L. M. Murrow, and Z. J. Gartner, “DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors,” Cell Systems 8, no. 4 (2019): 329–337.e4.

[36] V. Körber, S. A. Stainczyk, R. Kurilov, et al., “Neuroblastoma Arises in Early Fetal Development and its Evolutionary Duration Predicts Outcome,” Nature Genetics 55, no. 4 (2023): 619–630.

[37] J. A. Fresno Vara, E. Casado, J. de Castro, P. Cejas, C. Belda-Iniesta, and M. González-Barón, “PI3K/Akt Signalling Pathway and Cancer,” Cancer Treatment Reviews 30, no. 2 (2004): 193–204.

[38] A. S. Dhillon, S. Hagan, O. Rath, and W. Kolch, “MAP Kinase Signalling Pathways in Cancer,” Oncogene 26, no. 22 (2007): 3279–3290.

[39] S. W. Brady, Y. Liu, X. Ma, et al., “Pan-Neuroblastoma Analysis Reveals Age- and Signature-Associated Driver Alterations,” Nature Communications 11, no. 1 (2020): 5183.

[40] A. Zafar, W. Wang, G. Liu, et al., “Molecular Targeting Therapies for Neuroblastoma: Progress and Challenges,” Medicinal Research Reviews 41, no. 2 (2021): 961–1021.

[41] V. Mlakar, E. Morel, S. J. Mlakar, M. Ansari, and F. Gumy-Pause, “A Review of the Biological and Clinical Implications of RAS-MAPK Pathway Alterations in Neuroblastoma,” Journal of Experimental & Clinical Cancer Research 40, no. 1 (2021): 189.

[42] E. Secomandi, A. Salwa, C. Vidoni, A. Ferraresi, C. Follo, and C. Isidoro, “High Expression of the Lysosomal Protease Cathepsin D Confers Better Prognosis in Neuroblastoma Patients by Contrasting EGF-Induced Neuroblastoma Cell Growth,” International Journal of Molecular Sciences 23, no. 9 (2022): 4782.

[43] S. H. Lee, J. S. Kim, S. Zheng, et al., “ARID1B Alterations Identify Aggressive Tumors in Neuroblastoma,” Oncotarget 8, no. 28 (2017): 45943–45950.

[44] X. Li and P. M. Burgers, “Molecular Cloning and Expression of the Saccharomyces cerevisiae RFC3 Gene, an Essential Component of Replication Factor C,” Proceedings of the National Academy of Sciences of the United States of America 91, no. 3 (1994): 868–872.

[45] K. R. Avs, C. Pandi, B. Kannan, A. Pandi, V. P. Jayaseelan, and P. Arumugam, “RFC3 Serves as a Novel Prognostic Biomarker and Target for Head and Neck Squamous Cell Carcinoma,” Clinical Oral Investigations 27, no. 11 (2023): 6961–6969.

[46] S. Gong, X. Qu, S. Yang, S. Zhou, P. Li, and Q. Zhang, “RFC3 Induces Epithelial-Mesenchymal Transition in Lung Adenocarcinoma Cells through the Wnt/β-Catenin Pathway and Possesses Prognostic Value in Lung Adenocarcinoma,” International Journal of Molecular Medicine 44, no. 6 (2019): 2276–2288.

[47] M. Dahlhaus, A. Burkovski, F. Hertwig, et al., “Boolean Modeling Identifies Greatwall/MASTL as an Important Regulator in the AURKA Network of Neuroblastoma,” Cancer Letters 371, no. 1 (2016): 79–89.

[48] Y. Xu, K. Chen, Y. Cai, C. Cheng, Z. Zhang, and G. Xu, “Overexpression of Rad51 Predicts Poor Prognosis and Silencing of Rad51 Increases Chemo-Sensitivity to Doxorubicin in Neuroblastoma,” American Journal of Translational Research 11, no. 9 (2019): 5788–5799.

[49] K. A. Cole, J. Huggins, M. Laquaglia, et al., “RNAi Screen of the Protein Kinome Identifies Checkpoint Kinase 1 (CHK1) as a Therapeutic Target in Neuroblastoma,” Proceedings of the National Academy of Sciences of the United States of America 108, no. 8 (2011): 3336–3341.

[50] E. Fang, X. Wang, F. Yang, et al., “Therapeutic Targeting of MZF1-AS1/PARP1/E2F1 Axis Inhibits Proline Synthesis and Neuroblastoma Progression,” Advanced Science 6, no. 19 (2019): 1900581.

[51] O. Melaiu, M. Chierici, V. Lucarini, et al., “Cellular and Gene Signatures of Tumor-Infiltrating Dendritic Cells and Natural-Killer Cells Predict Prognosis of Neuroblastoma,” Nature Communications 11, no. 1 (2020): 5992.

[52] A. Garaventa, U. Poetschger, D. Valteau-Couanet, et al., “Randomized Trial of Two Induction Therapy Regimens for High-Risk Neuroblastoma: HR-NBL1.5 International Society of Pediatric Oncology European Neuroblastoma Group Study,” Journal of Clinical Oncology 39, no. 23 (2021): 2552–2563.

[53] D. Cangelosi, M. Morini, N. Zanardi, et al., “Hypoxia Predicts Poor Prognosis in Neuroblastoma Patients and Associates With Biological Mechanisms Involved in Telomerase Activation and Tumor Microenvironment Reprogramming,” Cancers 12, no. 9 (2020): 2343.

[54] A. Liang, Y. Kong, Z. Chen, et al., “Advancements and Applications of Single-Cell Multi-Omics Techniques in Cancer Research: Unveiling Heterogeneity and Paving the Way for Precision Therapeutics,” Biochemistry and Biophysics Reports 37 (2024): 101589.

[55] Y. Wu and K. Zhang, “Tools for the Analysis of High-Dimensional Single-Cell RNA Sequencing Data,” Nature Reviews Nephrology 16, no. 7 (2020): 408–421.

[56] Y. P. Mossé, M. Laudenslager, L. Longo, et al., “Identification of ALK as a Major Familial Neuroblastoma Predisposition Gene,” Nature 455, no. 7215 (2008): 930–935.

[57] S. Ackermann, M. Cartolano, B. Hero, et al., “A Mechanistic Classification of Clinical Phenotypes in Neuroblastoma,” Science 362, no. 6419 (2018): 1165–1170.

[58] J. A. Myers and J. S. Miller, “Exploring the NK Cell Platform for Cancer Immunotherapy,” Nature Reviews Clinical Oncology 18, no. 2 (2021): 85–100.

[59] J. Fan, S. Tang, X. Kong, and Y. Cun, “Integrating Multi-Omics Data Reveals Neuroblastoma Subtypes in the Tumor Microenvironment,” Life Sciences 359 (2024): 123236.

RIGHTS & PERMISSIONS

2025 The Author(s). Pediatric Discovery published by John Wiley & Sons Australia, Ltd on behalf of Children's Hospital of Chongqing Medical University.