Causal Structural Equation Modeling Links the Epigenetic Landscape to the Warburg Effect in Nasopharyngeal Carcinoma

Background: Nasopharyngeal carcinoma (NPC) is an Epstein-Barr virus (EBV)-associated epithelial malignancy characterized by profound epigenetic dysregulation and aerobic glycolysis (the Warburg effect). Although differential expression analyses consistently identify concurrent alterations across chromatin modifiers and glycolytic enzymes, standard correlative frameworks cannot test directional hierarchies between these biological hallmarks. Here, we applied structural equation modeling (SEM) to test whether the coordinated epigenetic landscape causally drives the Warburg effect in NPC. Methods: Laser-capture microdissected transcriptomic profiles from 41 samples (31 NPC tumors and 10 normal nasopharyngeal controls; NCBI GEO GSE12452 / GDS3341, Affymetrix HG-U133 Plus 2.0) were analyzed using limma, hierarchical clustering, and principal component analysis (PCA). Eleven differentially expressed biomarkers were assigned to two latent constructs: the Epigenetic Landscape (DNMT1, EZH2, HDAC2, SMARCA4, BRD4, and the long non-coding RNA H19) and the Warburg Effect (GAPDH, LDHA, CS, CYC, and TIMMDC1). Standardized transcriptomic scores were fitted via SEM in R (lavaan v0.6-21) using Satorra-Bentler robust maximum likelihood estimation. Results: All 11 indicators were significantly dysregulated in NPC (adjusted p < 0.001), featuring marked downregulation of H19 (log2FC = -1.8595) alongside coordinated upregulation of chromatin regulators and metabolic enzymes (log2FC = 0.4411 to 1.4546). PCA separated tumor from control tissues (73.7% cumulative variance) and resolved functional gene clusters along PC1 (94.4% variance). The structural model demonstrated acceptable global fit (scaled chi-square = 72.790, df = 41, p = 0.002; robust CFI = 0.915; SRMR = 0.080) with highly significant factor loadings (p < 0.001; H19 lambda = -0.861; positive indicators lambda = 0.655 to 0.942). Crucially, the Epigenetic Landscape exerted a strong, direct causal effect on the Warburg Effect (standardized beta = 0.805, p < 0.001), accounting for 64.8% of its latent variance (R² = 0.648). Conclusions: Epigenetic remodeling acts as an upstream driver of metabolic reprogramming in NPC, supporting dual epigenetic-metabolic therapeutic strategies.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-18
DOI
https://doi.org/10.21203/rs.3.rs-11767626/v1
Primary Topic
Head and Neck Cancer Studies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Causal Structural Equation Modeling Links the Epigenetic Landscape to the Warburg Effect in Nasopharyngeal Carcinoma

Antonio R. Romeu, Maria-Jose Oliach
2 links
Research Square
Head and Neck Cancer Studies
preprint

Causal Structural Equation Modeling Links the Epigenetic Landscape to the Warburg Effect in Nasopharyngeal Carcinoma

Antonio R. Romeu, Maria-Jose Oliach
preprint en

Abstract

Background: Nasopharyngeal carcinoma (NPC) is an Epstein-Barr virus (EBV)-associated epithelial malignancy characterized by profound epigenetic dysregulation and aerobic glycolysis (the Warburg effect). Although differential expression analyses consistently identify concurrent alterations across chromatin modifiers and glycolytic enzymes, standard correlative frameworks cannot test directional hierarchies between these biological hallmarks. Here, we applied structural equation modeling (SEM) to test whether the coordinated epigenetic landscape causally drives the Warburg effect in NPC. Methods: Laser-capture microdissected transcriptomic profiles from 41 samples (31 NPC tumors and 10 normal nasopharyngeal controls; NCBI GEO GSE12452 / GDS3341, Affymetrix HG-U133 Plus 2.0) were analyzed using limma, hierarchical clustering, and principal component analysis (PCA). Eleven differentially expressed biomarkers were assigned to two latent constructs: the Epigenetic Landscape (DNMT1, EZH2, HDAC2, SMARCA4, BRD4, and the long non-coding RNA H19) and the Warburg Effect (GAPDH, LDHA, CS, CYC, and TIMMDC1). Standardized transcriptomic scores were fitted via SEM in R (lavaan v0.6-21) using Satorra-Bentler robust maximum likelihood estimation. Results: All 11 indicators were significantly dysregulated in NPC (adjusted p < 0.001), featuring marked downregulation of H19 (log2FC = -1.8595) alongside coordinated upregulation of chromatin regulators and metabolic enzymes (log2FC = 0.4411 to 1.4546). PCA separated tumor from control tissues (73.7% cumulative variance) and resolved functional gene clusters along PC1 (94.4% variance). The structural model demonstrated acceptable global fit (scaled chi-square = 72.790, df = 41, p = 0.002; robust CFI = 0.915; SRMR = 0.080) with highly significant factor loadings (p < 0.001; H19 lambda = -0.861; positive indicators lambda = 0.655 to 0.942). Crucially, the Epigenetic Landscape exerted a strong, direct causal effect on the Warburg Effect (standardized beta = 0.805, p < 0.001), accounting for 64.8% of its latent variance (R² = 0.648). Conclusions: Epigenetic remodeling acts as an upstream driver of metabolic reprogramming in NPC, supporting dual epigenetic-metabolic therapeutic strategies.

Research Square
Universitat Rovira i Virgili (ES)
Good health and well-being
Head and Neck Cancer Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon