Network Analysis Pinpoints Five Hub Genes Steering Hodgkin Lymphoma Pathways

Hodgkin Lymphoma has long presented researchers with a paradox: contemporary chemotherapy and immunotherapy cure a large majority of patients, yet a stubborn fraction relapse or become refractory to treatment, and the molecular machinery driving progression, immune evasion and therapeutic resistance remains only partially mapped. A new integrative bioinformatics study, published in the Egyptian Journal of Medical Human Genetics, tackles that gap not with new laboratory experiments but with a large-scale computational sweep through public gene-disease databases, and it arrives at a compact shortlist of five genes that appear to sit at the centre of the disease’s molecular wiring: TP53, STAT3, AKT1, EGFR and IL6.

The research team, led by S. B. Ankitha and colleagues, began by harvesting Hodgkin Lymphoma-associated genes from two independent public resources. A search of GeneCards, restricted to genes with a relevance score above 10, returned 3,915 candidate genes. The Comparative Toxicogenomics Database, which curates gene-disease-chemical interactions drawn from toxicological and biomedical research, contributed a far broader list of 22,171 genes tied to Hodgkin Lymphoma and related biological mechanisms. By standardising gene identifiers and intersecting the two sets, the authors retained only genes supported by both independent sources, yielding 3,180 overlapping genes that they treated as a high-confidence disease-associated dataset.

That intersection strategy matters because database annotations differ in origin and reliability. Genes appearing in both GeneCards, an integrative annotation platform, and the manually curated CTD carry two lines of evidence for their association with the disease, reducing the false-positive burden that plagues single-source screens. The 3,180-gene set then became the input for every downstream analysis: pathway enrichment, protein-protein interaction mapping and, ultimately, hub gene prioritisation.

To ask what these genes collectively do, the team uploaded the full list to DAVID, the Database for Annotation, Visualization and Integrated Discovery, and ran enrichment analysis against the Kyoto Encyclopedia of Genes and Genomes. Statistical significance was assessed with Benjamini-Hochberg false discovery rate correction, with an adjusted p-value below 0.05 considered significant. The standout result came from the KEGG category labelled Pathways in Cancer: 295 of the overlapping genes were significantly enriched there, indicating that a substantial share of the HL-associated gene set participates in oncogenic signalling cascades governing cell proliferation, apoptosis, immune responses, inflammatory signalling and cellular stress.

The next step was structural. The researchers submitted all 3,180 genes to STRING, the Search Tool for the Retrieval of Interacting Genes/Proteins, applying a confidence score threshold of at least 0.700 so that only high-confidence interactions entered the network. STRING integrates experimentally validated interactions with computationally predicted functional associations drawn from curated databases, co-expression data and literature mining. The resulting protein-protein interaction network, visualised and analysed in Cytoscape, revealed a densely interconnected molecular web punctuated by several tightly clustered modules, which the authors interpret as functional units coordinating tumour progression, immune regulation and cellular signalling.

Hub identification relied on a simple but widely used topological metric: degree centrality, which counts the number of direct connections each node maintains. Nodes with the highest connectivity often act as major regulatory points in biological networks, and in this analysis five genes rose to the top. TP53, the best-studied tumour suppressor in cancer biology, emerged as the leading hub, followed by the transcription factor STAT3, the survival kinase AKT1, the growth factor receptor EGFR and the inflammatory cytokine IL6. All five are associated with processes central to lymphoma biology, including apoptosis regulation, immune signalling and cellular proliferation, and their extensive interactions with surrounding nodes suggest roles as central coordinators rather than peripheral participants.

Each of the five carries independent biological weight in the lymphoma literature, as the authors review in their discussion. TP53 inactivation impairs DNA damage responses in Hodgkin Lymphoma and has been linked to chemoresistance, and MDM2 inhibitors designed to restore p53 activity are under clinical investigation in lymphoid malignancies. Constitutive STAT3 signalling promotes proliferation, survival and immune evasion while fostering an immunosuppressive tumour microenvironment, and JAK/STAT inhibitors, alone or combined with immune checkpoint blockade, are being explored for relapsed or refractory disease. AKT1 anchors the PI3K/AKT/mTOR axis, whose pharmacological inhibition has shown activity in preclinical lymphoma models and is being tested clinically. EGFR overexpression has been detected in Reed-Sternberg cells, the malignant hallmark of HL, although EGFR-targeted therapy is not part of standard HL management. Elevated IL6 levels have been associated with adverse prognosis in HL patients.

The study did not stop at network topology. To probe expression patterns, the team examined publicly available expression resources, reporting that the five hub genes show varying levels across different malignancies and disease stages, and that enrichment analysis also linked several of them to diffuse large B-cell lymphoma, a related but clinically distinct lymphoid cancer. That molecular overlap, the authors suggest, points to oncogenic and immune-regulatory pathways conserved across lymphoma subtypes, echoing previously reported alterations of the same genes in DLBCL and chronic lymphocytic leukemia. A separate survival analysis, built on publicly available cancer datasets rather than HL-specific patient cohorts, found that alterations in these genes were associated with variations in patient survival outcomes across multiple cancer types, an exploratory observation the authors themselves flag as requiring validation in HL-specific cohorts.

Independent transcriptomic support came from the Gene Expression Omnibus dataset GSE12453, which contains microarray profiles of Hodgkin Lymphoma cell lines and normal B-cell controls generated on the Affymetrix Human Genome U133 Plus 2.0 platform. Using the GEO2R tool to compare predefined groups, the researchers found that STAT3, AKT1 and IL6 showed elevated expression in HL samples relative to normal B cells, while TP53 and EGFR displayed variable expression patterns across samples. The authors present this as transcriptomic corroboration of the hub genes’ involvement in HL-associated molecular pathways, while noting that cell-line data cannot substitute for patient tissue.

The authors are candid about the limits of their approach. The entire analysis rests on publicly available databases, inheriting their annotation biases and heterogeneity; the STRING network pools interaction evidence from many biological contexts and therefore cannot be assumed to represent interactions specific to Hodgkin Lymphoma; and the absence of patient-level transcriptomic datasets prevents any direct correlation between gene expression and clinical outcomes in HL. The five hub genes, in other words, are hypotheses rather than validated biomarkers, and the authors state that experimental validation in HL-specific clinical cohorts is required before any translation into practice. What the study delivers is a systems-level map: a demonstration that the molecular architecture of Hodgkin Lymphoma, as reflected in two independent gene-disease databases, converges on a small set of highly connected regulatory nodes within cancer-associated signalling pathways, and a prioritised shortlist for the functional experiments that must come next.

Subject of Research: Integrative bioinformatics network analysis of Hodgkin Lymphoma identifying hub genes and cancer-associated pathways

Article Title: Functional network analysis of Hodgkin Lymphoma identifies critical regulatory nodes in cancer-associated pathway

Article References: Ankitha, S. B., Krishnasamy, N., Selvam, S. P., Ramadoss, R., & Sundar, S. (2026). Functional network analysis of Hodgkin Lymphoma identifies critical regulatory nodes in cancer-associated pathway. Egyptian Journal of Medical Human Genetics, 27(1), Article 75. https://doi.org/10.1186/s43042-026-00904-x

Image Credits: AI Generated

DOI: 10.1186/s43042-026-00904-x

Keywords: Hodgkin Lymphoma, bioinformatics, hub genes, protein-protein interaction network, TP53, STAT3, AKT1, EGFR, IL6, KEGG pathway enrichment, GeneCards, STRING database