Abstract <p>Head and neck squamous cell carcinoma (HNSCC) sometimes co-occurs with other disorders, although their molecular causes remain unknown. This study aims to computationally reveal HNSCC comorbidities via molecular and functional connections between them using a biological multiplex network framework. The proposed method provides an insight into disease’s systemic landscape by a multi-layered graph with machine learning application. A multiplex network was formed by combining gene expression correlation, physical interactions, and similar molecular pathways. Louvain RWR clustering identifies the network’s functional modules, ranking them by disease and functional enrichment, as well as literature evidence. TCGA-HNSCC clinical data linked enriched diseases to anatomical sites, validated with metastatic tumor locations. The proposed framework identified 75 functional modules, of which 68 satisfied validation criteria. These modules captured most known HNSCC comorbidity categories, including 11 of 16 reported by the Morbinet clinical analysis, confirming methodological concordance. Validation using an independent cohort, CPTAC reproduced highly similar functional and disease patterns, demonstrating cross-cohort reproducibility. Network robustness analyses further confirmed that module structure and disease enrichment remained stable under null models and perturbations. In addition to known associations, the framework revealed novel comorbidities within established disease categories, highlighting new directions for investigation. This research highlights the effectiveness of a multiplex network framework in computationally uncovering the functional and molecular basis of HNSCC comorbidities. By providing a fine-grained view of established comorbidity categories, the results improve understanding of HNSCC’s systemic relationships and lay the foundation for future clinical and molecular validation toward personalized therapeutic strategies.</p>